In the ever-evolving landscape of data analytics and business intelligence, Microsoft continues to push boundaries and empower developers with cutting-edge tools. The latest breakthrough comes in the form of Microsoft Fabric, a powerful integration that revolutionizes collaboration within development teams working on Power BI content.

With the integration of Git, Microsoft Fabric brings a new level of efficiency and cohesion to the development process. Now, connecting your Power BI workspace to Azure DevOps repositories is a breeze, unlocking a multitude of benefits that streamline development, version control, and deployment.

Gone are the days of scattered changes, conflicting updates, and a lack of synchronization. Microsoft Fabric enables developers to effortlessly merge updates from multiple team members, transforming a myriad of modifications into a single source of truth. And with a single click, this unified version is seamlessly synced into the workspace, ensuring everyone is always on the same page.

As a developer, this integration provides a wealth of opportunities to enhance your Power BI development workflow. From authoring report and dataset metadata files in source-control friendly formats to enabling code reviews and automated testing, Microsoft Fabric empowers you to take full control of your development process. By leveraging continuous integration and continuous delivery (CI/CD), quality gates can be enforced, ensuring only the most refined and validated content reaches production environments.

But the benefits extend beyond just version control and development practices. Microsoft Fabric equips users with the tools to manage their entire application lifecycle seamlessly. Whether it’s leveraging Git integration or deploying Power BI content across dev, test, and production workspaces, this integration provides a holistic approach to application management. Developers can choose to utilize the intuitive user interface (UI) experience or automate the process through other tools like Azure Pipelines, providing flexibility to match individual preferences.

In this Medium post, we will delve into the details of the git Integration between Power BI and Azure DevOps repositories, exploring the diverse capabilities it offers and how it can transform your Power BI development experience. From empowering collaboration to ensuring the integrity of deployments, this integration is set to redefine the way development teams collaborate and deliver exceptional Power BI solutions.

## How to integrate Power BI with git?

Now, with Power BI Desktop, you not only have the option to save your work as a single PBIX file, but you can also save it as a Power BI project (.PBIP).

From an open Power BI reports, you can now go to File Save As, and select the Power BI projects Save As type, and Power BI desktop will save all your developments into a folder.

![](https://miro.medium.com/v2/resize:fit:700/1*ikZ2XFul9HZmsFsWYH_dZg.png)

This is a folder with a Power BI project in it, it contains one folder for the dataset, and one folder for the report.

![](https://miro.medium.com/v2/resize:fit:700/1*wg35el5XormKNsaLcOWBkA.png)

If you go back to the desktop and you create a new measure, desktop will save the new measure in the model definition within the dataset folder.

![](https://miro.medium.com/v2/resize:fit:700/1*_6q_lOirpzEf5LA52bsVrg.png)

You can initialize a Git repo and enable Version Control and collaboration with other developers. You can do that using Visual Studio Code. From the Power BI project folder, you can open Visual Studio Code and initialize and you’ll get a repo.

![](https://miro.medium.com/v2/resize:fit:700/1*anPtDtb2_s274ZPb7kSblw.png)![](https://miro.medium.com/v2/resize:fit:668/1*wWweuU-fr-NXs0l8BfAkHQ.png)

From now on, because you have a Git Enabled folder, you can track and version control any change you make in desktop, for example, if you change a measure, you can easily track that change in Git. Visual Studio Code will show you that you have a file div in your model div file.

![](https://miro.medium.com/v2/resize:fit:700/1*jeuSKugJliZAp54SANawvw.png)

Now, to enable collaboration, you need to use Azure DevOps. You can go to Azure DevOps and you can create a new repo.

![](https://miro.medium.com/v2/resize:fit:700/1*FO5vJqcwRNN3s23gCPjmZQ.png)![](https://miro.medium.com/v2/resize:fit:700/1*zJItpTjiZBK4pA5dKLqd_A.png)

You need to configure this remote Git repo URL back in Visual Studio Code and publish your branch.

![](https://miro.medium.com/v2/resize:fit:700/1*yXjSbGNQF_9GIOx2U-2MXw.png)![](https://miro.medium.com/v2/resize:fit:383/1*VYXUEbNJWUYENtpAz1PtWQ.png)

This Visual Studio Code will take care of sinking your local development into Azure DevOps.

![](https://miro.medium.com/v2/resize:fit:700/1*BF3QXB8qhsgylTRMeMVobQ.png)

Here now you can enable collaboration and have multiple developers working on the same Power BI project using Power BI desktop, they only need to be connected to the same Azure DevOps repo.

But we didn’t stop here. You can also enable to sink a Git repo to a workspace in the service.

For that, you need to go to the workspace settings. Where you will find an option called Git integration that will allow you to connect the workspace to an Azure DevOps Git repo. Let’s select the Azure DevOps organization and the projects, the repo, and the branch we were working and click on Connect and Sink.

![](https://miro.medium.com/v2/resize:fit:700/1*DYThAloQqCs0mOKHLN3R2w.png)

Just like that, we just enable that two-way synchronization between the workspace and a Git repo in Azure DevOps, it will start by synchronizing the content from Git into the workspace, creating a report and a dataset artifact.

## Conclusion

As we conclude our exploration of Microsoft Fabric’s game-changing capabilities, it’s evident that this integration has truly ushered in a new era of seamless collaboration and streamlined development for Power BI content.

The ability to author report and dataset metadata files in source-control friendly formats, enable multiple developer collaboration, and enforce quality gates through continuous integration and continuous delivery (CI/CD) are just a few examples of the immense value Microsoft Fabric brings to the table.

Moreover, the integration with Azure DevOps repositories and the utilization of Git integration and deployment pipelines enable developers to have complete control over the application lifecycle. From development to testing and production, Microsoft Fabric provides a robust framework for end-to-end application management.

The future is here, and it’s powered by Microsoft Fabric.

References:

https://powerbi.microsoft.com/en-us/blog/introducing-microsoft-fabric-and-copilot-in-microsoft-power-bi/

https://youtu.be/OdkS7DF7ElY?t=684


Git Integration with Power BI

In this digital age, where information overload is the norm, it has become increasingly crucial to present data in a way that captures attention, conveys insights, and drives meaningful action.

Visuals possess an incredible ability to transform complex datasets into meaningful stories, making them accessible and engaging to a wide audience. However, with an array of visual options at your disposal, it’s essential to understand the strengths and weaknesses of each type to make an informed decision. In this article, we will explore twelve key visualizations: simple text, scatterplot, table, line chart, heat map, slope graph, vertical bar, horizontal bar, stacked vertical bar, stacked horizontal bar, waterfall, and square area. By understanding the purpose and characteristics of each, you will gain valuable insights into when and how to use them effectively.

Whether you’re a data analyst, business professional, or simply someone interested in enhancing your data communication skills, this blog post will equip you with the knowledge and tools to confidently select the most suitable visual representation for your data. Follow me on Medium, where I share insightful articles and valuable tips on harnessing the power of data visualization to make informed decisions. Let’s dive in!

**Simple text**

If you have only one or a few numbers to share, using simple text can be an effective method of communication. It’s important to emphasize the number itself and provide a few supporting words to convey your message clearly. Simply placing one or a couple of numbers in a table or graph can potentially be misleading and diminish their impact. When you need to communicate one or two numbers, it’s best to rely on the numbers themselves.

![](https://cdn-images-1.medium.com/max/800/1*Xzplpynip4q9nC2bbT5YuA.png)![](https://cdn-images-1.medium.com/max/800/1*AZYpQ10e6dBQrDID_KkqIg.png)

When you only have one or a few numbers to communicate, it’s best to present them directly. However, when dealing with larger sets of data, tables or graphs are generally more suitable. One thing to understand is that people interact differently with these two types of visuals. Let’s discuss each in detail and look at some specific varieties and use cases.

**Tables**

Tables are designed to be read as part of our verbal system. When we encounter a table, we tend to visually scan across rows and down columns or compare values. Tables excel at communicating information to diverse audiences, allowing each person to easily locate their specific row of interest. Furthermore, if you need to present multiple units of measurement, tables typically offer a more straightforward approach compared to graphs.

**PRO TIP:** Using a table during a live presentation is generally not recommended. When the audience focuses on reading the table, their attention shifts away from your verbal message. When incorporating a table into a presentation or report, it is important to consider the main point you are trying to convey. In most cases, there are better ways to highlight and visualize the relevant information. If you feel that omitting the entire table would result in significant loss, consider placing it in the appendix and providing a reference or link for your audience’s convenience.

When designing a table, it is crucial to ensure that the visual elements do not distract from the data itself. Avoid using heavy borders or shading that may compete for attention. Instead, opt for light borders or utilize white space to distinguish different elements within the table.

Borders in a table should serve the purpose of enhancing legibility. Consider making them grey or eliminating them altogether to allow the data to take center stage. The focus should be on the information presented, rather than the borders themselves.

**Heatmap**

One approach for mixing the detail you can include in a table while also making use of visual cues is via a heatmap. A heatmap is a way to visualize data in tabular format, where in place of (or in addition to) the numbers, you leverage colored cells that convey the relative magnitude of the numbers.

![](https://cdn-images-1.medium.com/max/800/1*Ho7R91-IxEI-UKjy_rGQ3w.png)![](https://cdn-images-1.medium.com/max/800/1*ZMjnD1ATTHErRztpZ8_Hrw.png)

**Color saturation **provides visual cues, helping our eyes and brains more quickly target the potential points of interest. On the Heatmap, the higher saturation of blue, the higher the number. This makes the process of picking out the tails of the spectrum — the lowest number (0.2) and highest number (2.25) — an easier and faster process to direct our attention.

**Graphs**

Although tables engage our verbal system, graphs engage our visual system, which is known for its faster information processing. Consequently, a well-designed graph generally conveys information more quickly than a well-designed table. As mentioned earlier in this chapter, there is an abundance of graph types available. The positive aspect is that a few of these graph types will suffice for most common requirements.

The types of graphs experts frequently use fall into four categories: points, lines, bars, and area. We will examine these more closely and discuss the subtypes that experts find themself using on a regular basis, with specific use cases and examples for each.

#### **Points**

**Scatterplot**

Scatterplots serve as a valuable tool for illustrating the relationship between two variables by representing data simultaneously on both the horizontal x-axis and the vertical y-axis. They are commonly employed in scientific disciplines to examine relationships, although they may appear complex to individuals less acquainted with them. Though infrequent, there are use cases for scatterplots in the business world as well.

![](https://cdn-images-1.medium.com/max/800/1*JwUV9hvB11cGJVE_kYQZcw.png)![](https://cdn-images-1.medium.com/max/800/1*eogqZNaBgEMnrylahKXLDg.png)

#### **Lines**

Line graphs are frequently employed for representing continuous data. The connected line between the points implies a relationship that may not be meaningful for categorical data, which consists of data sorted or divided into different categories. Continuous data is often measured over time, such as days, months, quarters, or years.

Within the line graph category, there are two types of charts that data visualization specialist frequently find themself using: the standard line graph and the slopegraph.

**Line graph**

The line graph can show a single series of data, two series of data, or multiple series

![](https://cdn-images-1.medium.com/max/800/1*vwo15Y2c4IpvrbdcT5Tl4A.png)![](https://cdn-images-1.medium.com/max/800/1*q_-w2jV2o2ZUvl0IVzZs4g.png)

**PRO TIP:** In some cases, the line in your line graph may represent a summary statistic, like the average, or the point estimate of a forecast. If you also want to give a sense of the range (or confidence level, depending on the situation), you can do that directly on the graph by also visualizing this range.

**Slopegraph**

Slopegraphs can be useful when you have two time periods or points of comparison and want to quickly show relative increases and decreases or differences across various categories between the two data points.

![](https://cdn-images-1.medium.com/max/800/1*8OYMjGI3R-d1ihvUJiTuFA.png)![](https://cdn-images-1.medium.com/max/800/1*fGgqFBxajkTwimnXRhnT9A.png)

Slopegraphs pack in a lot of information. In addition to the absolute values (the points), the lines that connect them give you the visual increase or decrease in rate of change (via the slope or direction) without ever having to explain that’s what they are doing, or what exactly a “rate of change” is — rather, it’s intuitive.

**PRO TIP:** Whether a slopegraph will work in your specific situation depends on the data itself. If many of the lines are overlapping, a slopegraph may not work, though in some cases you can still emphasize a single series at a time with success. For example, we can draw attention to the single category that decreased over time from the preceding example.

#### **Bars**

Bar charts should not be avoided simply because they are common. In fact, their familiarity makes them advantageous as there is less learning required for the audience. Bar charts are easy to read, allowing for quick comparison of categories based on the endpoints of the bars. It is important to ensure that bar charts have a zero baseline to avoid misleading visual comparisons. 

**PRO TIP:** When graphing data, a common decision to make is whether to preserve the axis labels or eliminate the axis and instead label the data points directly. In making this decision, consider the level of specificity needed. If you want your audience to focus on big‐picture trends, think about preserving the axis but deemphasizing it by making it grey.
If the specific numerical values are important, it may be better to label the data points directly. In this latter case, it’s usually best to omit the axis to avoid the inclusion of redundant information. Always consider how you want your audience to use the visual and construct it accordingly.

While we’re considering lengths of bars, let’s also spend a moment on the width of bars. There’s no hard‐and‐fast rule here, but in general the bars should be wider than the white space between the bars. You don’t want the bars to be so wide, however, that your audience wants to compare areas instead of lengths. 

**Vertical bar chart**

The basic type of bar chart is the vertical bar chart, also known as a column chart. Similar to line graphs, vertical bar charts can consist of a single series, two series, or multiple series of data. However, it is important to exercise caution when using multiple series bar charts as it can become challenging to focus on and derive insights from each series individually. Additionally, the spacing in bar charts with multiple data series results in visual grouping, making the relative order of categorization significant. When creating a bar chart, carefully consider the desired comparisons and structure the categorization hierarchy to facilitate easy understanding for the audience.

![](https://cdn-images-1.medium.com/max/800/1*mKuPFvg2vgdohPgOxw2sqg.png)![](https://cdn-images-1.medium.com/max/800/1*EAKG5A4s7j_EvevIjYQEgg.png)

**Stacked vertical bar chart**

The usage of stacked vertical bar charts is relatively limited. They serve the purpose of comparing totals across categories while also providing visibility of the subcomponents within each category. However, using stacked vertical bar charts can quickly become visually overwhelming, particularly due to the varied default color schemes in most graphing applications. It becomes challenging to compare the subcomponents across different categories, especially beyond the bottom series adjacent to the x-axis, as there is no consistent baseline for comparison. This complicates the comparison for our visual perception.

![](https://cdn-images-1.medium.com/max/800/1*w-ujegywTPy65uJwyZymZw.png)![](https://cdn-images-1.medium.com/max/800/1*-U1j6VylJ_1weq5xnfaLrA.png)

**Waterfall chart**

The waterfall chart can be used to pull apart the pieces of a stacked bar chart to focus on one at a time, or to show a starting point, increases and decreases, and the resulting ending point.

![](https://cdn-images-1.medium.com/max/800/1*bNvA1Sqv6XLXCgHC1re6UA.jpeg)

**Horizontal bar chart**

If I had to choose a single preferred graph for categorical data, it would be the horizontal bar chart, which is a flipped version of the vertical bar chart. The reason for this choice is its exceptional readability. The horizontal bar chart is particularly beneficial when dealing with long category names, as the text is aligned from left to right, following the natural reading pattern of most audiences, ensuring legibility. Additionally, considering our typical information processing pattern, where our eyes start from the top left and move in a zigzag pattern across the screen or page, the structure of the horizontal bar chart allows us to encounter the category names before the actual data. This enables us to understand what the data represents by the time we reach it, eliminating the need for constant eye movements between the data and category names, as required in vertical bar charts.

Like the vertical bar chart, the horizontal bar chart can be single series, two series, or multiple series

![](https://cdn-images-1.medium.com/max/800/1*y8YOO_xpnqW-ViUuHPGUyw.png)![](https://cdn-images-1.medium.com/max/800/1*b44EcHMNlmfMIv8PVGJ1aA.png)

**PRO TIP:** When designing any graph showing categorical data,
be thoughtful about how your categories are ordered. If there is a natural ordering to your categories, it may make sense to leverage that. For example, if your categories are age groups — 0–10 years old, 11–20 years old, and so on — keep the categories in numerical order. If, however, there
isn’t a natural ordering in your categories that makes sense to leverage, think about what ordering of your data will make the most sense. Being thoughtful here can mean providing a construct for your audience, easing the interpretation process.

Your audience (without other visual cues) will typically look
at your visual starting at the top left and zigzagging in “z” shapes. This means they will encounter the top of your graph first. If the biggest category is the most important, think about putting that first and ordering the rest of the categories in decreasing numerical order. Or if the smallest is most important, put that at the top and order by ascending data values.

**Stacked horizontal bar chart**

Just like the stacked vertical bar chart, stacked horizontal bar charts can be employed to display totals across various categories while also providing insights into the subcomponents. These charts can be organized to present either absolute values or proportions that sum up to 100%. 

Experts have found that the latter approach is effective for representing portions of a whole on a scale ranging from negative to positive. This is because the consistent baselines on both the far left and far right facilitate straightforward comparisons of both the leftmost and rightmost components.

![](https://cdn-images-1.medium.com/max/800/1*b5JNpQwSA1UoPZvRTEe13g.png)![](https://cdn-images-1.medium.com/max/800/1*IJNB4_g_nn21qhrSFWkPNQ.png)

**Area**

Specialist generally steer clear of using area graphs. Human perception struggles to accurately assign numerical values to two-dimensional space, making area graphs more challenging to interpret compared to other visual displays we have discussed. As a result, specialists typically avoid them, except for one scenario: when there is a need to visualize numbers that vary significantly in magnitude. The additional dimension provided by using squares (which have both height and width, unlike bars that have only one dimension) enables a more concise representation of such data compared to using a single dimension.

![](https://cdn-images-1.medium.com/max/800/1*0atfDcDB_N_U8EnMh5phlg.png)![](https://cdn-images-1.medium.com/max/800/1*dUazK6q8ZHVTGCzQihug_w.png)

**To be avoided**

We’ve discussed the visuals that Data Visualization Specialist use most commonly to communicate data in a business setting. There are also some specific graph types and elements that you should avoid: pie charts, donut charts, 3D, and secondary y‐axes. Let’s discuss each of these.

**Pie charts are evil**

Pie charts are strongly criticized due to several issues they pose for accurate data interpretation. The application of 3D and strange perspectives in pie charts distorts the perception of the sizes of the pie pieces, making some appear smaller or larger than they actually are. Even when the 3D effect is removed, challenges in interpreting the data persist. The human eye struggles to assign quantitative value to two-dimensional space, making it difficult to determine the relative sizes of close segments and quantify the differences between larger segments. Although data labels can be added to aid interpretation, some argue that the visual representation of pie charts is not worth the space it occupies.

**Never use 3D**

One of the fundamental principles of data visualization emphasizes avoiding the use of 3D, with few exceptions. Adding a third dimension can complicate data interpretation and comparisons. In the case of a single dimension, using 3D distorts the numbers and hinders their understanding. The inclusion of 3D elements in graphs introduces unnecessary distractions such as side and floor panels. Graphing applications often exhibit peculiar behavior when plotting values in 3D, making it even more challenging to interpret the visual representation accurately.

**Secondary y‐axis: generally not a good idea**

There are instances where plotting data with different units on the same x-axis is useful. This situation often leads to the consideration of using a secondary y-axis, positioned on the right-hand side of the graph. However, using a secondary y-axis can cause confusion and hinder data interpretation. Instead, two alternative approaches can be employed: labeling the data points directly without showing the secondary axis, or separating the graphs vertically with separate y-axes on the left side while maintaining the same x-axis for both.

**Finally**

In conclusion, choosing an effective visualization is crucial for effectively communicating data and insights. By considering the nature of the data, the intended message, and the target audience, one can make informed decisions on which visualization type to use. It is essential to align the chosen visualization with the data’s characteristics and leverage the strengths of each type to enhance clarity and understanding. Additionally, avoiding common pitfalls such as using excessive 3D effects or cluttering the visual with unnecessary elements can significantly improve the effectiveness of the chosen visualization. Ultimately, selecting the right visualization is an art that requires thoughtful consideration and practice, but it can greatly enhance the impact and persuasive power of data-driven narratives.

Reference: 

<http://www.storytellingwithdata.com/book/>

<https://github.com/adamribaudo/storytelling-with-data-ggplot>

<https://github.com/empathy87/storytelling-with-data>











































The magic of choosing an EFFECTIVE visual

Shiny for Python is now available in the alpha stage and this article will guide you in the usage of the framework. Furthermore, I use this post to introduce some knowledge of key financial indicators that are used in value investing. The idea is to be able to research companies with competitive advantages using only six key indicators that are used by the well-known fathers of Value Investing like Warren Buffet and Ben Graham.

> Shiny for Python and Shiny in R are almost identical

I will explain each of the indicators plus I will share step-by-step the code that was used in this little project. But first of all, I need to clarify that you need to have access to [Financial Model Prep API (FMP)](https://site.financialmodelingprep.com). You should create an account or log in to the website if you are looking to follow the steps explained in this post or, if you wish, you could just modify this dashboard and create a new one with annual data that is free with limited numbers of API calls in FMP. The project could be cloned from this URL: <https://github.com/octadelsueldo/ValueInvestingDashboard>

The goal of the investment dashboard, following a value investment philosophy, is to find:

*   Profitable Companies

*   With low Debt

*   That provides value to their investors (positive EVA)

*   Without the risk of bankruptcy (Altman Z-Score)

*   Without earnings manipulation (Beneish Score)

# What metrics should we use?

## Operative Margin

> Is this company capable of consistently generating a positive operative margin in their business activities?

The operating margin measures how much profit a company makes on a dollar of sales after paying for variable costs of production, such as wages and raw materials, but before paying interest or tax. It is calculated by dividing a company’s operating income by its net sales. Higher ratios are generally better, illustrating the company is efficient in its operations and is good at turning sales into profits. 

We want to invest in companies that are capable of growing, so here we will look forward to checking the slope of the operative margin. We need to reach a positive slope to approve the company as profitable over the years.

## Debt rate

> Total debt relative to the amount originally invested by the owners

The debt-to-equity ratio measures your company’s total debt relative to the amount originally invested by the owners and the earnings that have been retained over time. Bear in mind that in this case, we are looking for a debt-to-equity rate that is below 0.5 to reach our target

## Retained Earnings

> Retained earnings (RE) are the amount of net income left over for the business after it has paid out dividends to its shareholders.

Retained earnings are the cumulative net earnings or profits of a company after accounting for dividend payments. As an important concept in accounting, the word “retained” captures the fact that because those earnings were not paid out to shareholders as dividends, they were instead retained by the company.

For this reason, retained earnings decrease when a company either loses money or pays dividends and increase when new profits are created. To pass the value investing test of retained earnings, here we need a company that has a power of the slope of retained earnings above 0. This is the measure we want to reach.

## EVA

> EVA shows how and where a company created wealth, through the inclusion of balance sheet items.

EVA is the incremental difference in the rate of return (RoR) over a company's cost of capital. Essentially, it is used to measure the value a company generates from funds invested in it. If a company's EVA is negative, it means the company is not generating value from the funds invested into the business. Conversely, a positive EVA shows a company is producing value from the funds invested in it. In this case, we are looking for companies that provide value to their investors over the last 10 years. So despite the fact that we need to check the last period as the most important, we need to revise the whole period with positive EVA.

The formula for calculating EVA is:

EVA = NOPAT - (Invested Capital \* WACC)

## Altman Z-Score

> The formula takes into account profitability, leverage, liquidity, solvency, and activity ratios. 

The Altman Z-score is the output of a credit-strength test that gauges a publicly traded manufacturing company's likelihood of bankruptcy.

The formula takes into account profitability, leverage, liquidity, solvency, and activity ratios. An Altman Z-score close to 0 suggests a company might be headed for bankruptcy, while a score closer to 3 suggests a company is in a solid financial position.

## Beneish Score

> if a company scored greater than -2.22 (i.e. a less negative or positive number) there was a likely probability of profit manipulation.

Beneish’s M-Score is a mathematical model that uses eight financial ratios weighted by coefficients to identify whether a company has manipulated its profits. It was created by Professor Messod Beneish who published a paper in June 1999 called The Detection of Earnings Manipulation. Beneish surmises that companies are incentivized to manipulate profits if they have high sales growth, deteriorating gross margins, rising operating expenses, and rising leverage. They are likely to manipulate profits by accelerating sales recognition, increasing cost deferrals, raising accruals, and reducing depreciation. 

Beneish concluded that if a company scored greater than -2.22 (i.e. a less negative or positive number) there was a likely probability of profit manipulation.

# The Web App

The panel sidebar consists of two text inputs for the FMP API key, the ticker, and one radio button input to choose the number of quarters to be retrieved.

The main panel consists of 6 plots. First of all the price over time shows the tendency of the stock price. Next we have the Key financial indicators presented in different rows and columns side by side. Here, down below, you can see a block of code with the first lines of the Shiny App.



On the side of the server (where magic happen), we will be calling the FMP API to call the financial statements such as the income statement, balance sheet, different financial ratios, etc. Additionally, I added a reactive values that will be updated each time we press the buttom input "input_action_button". This reactive values will help us when we would like to update the plots for a different stock. I mean this object will be converted in a reactive dataset that are updated each time we update the button, so when they are updated, our plot understand that the stock just changed and the reactive dataset was updated so the plots will changed accordingly.



To conclude
You should now be able to go through your favorite stocks and rapidly get a sense of their value indicators to make an investment decision with fundamentals.

The metrics were selected based on value investing books and courses , but you should be able to replace them by others of your preference. I also recommend you to use a screener like yahoo finance, zacks u other to make a first filter companies based on their key indicators before you get into the shiny app to research the last 10 years.

I'’ll be adding new shiny apps with additional charts and test to go deep in the value investing analysis in my next posts. Please feel free to tell me a suggestion or a feedback about the post.

References:

https://www.investopedia.com/

https://www.bdc.ca/

https://www.gmtresearch.com/en/accounting-ratio/beneishs-m-score/

If you are new to Medium you can join by following this link.

A small part of your membership will support my writing and the other talented writers on this platform.

Originally published at https://www.octaviods.com on October 10, 2022.


Shiny Web App to visualize key financial metrics

firstName

lastName

email

updatesConsent

Stay up-to-date with my words ✍️

Exploratory data analysis (EDA) is frequently used by data scientists or data analysts to comprehend datasets for use in decision-making and data cleansing procedures. Important facts about the data, including hidden patterns, outliers, variance, covariance, and correlations between characteristics, are revealed by EDA. The data is crucial for designing the hypothesis and building more effective models.

EDA often fits into one of two categories, univariate and multivariate analysis. Analyzing one feature, such as summarizing and identifying feature patterns, is what univariate analysis entails. The univariate exploratory data analysis works with four summary measures:

*   Position

*   Dispersion

*   Shape

*   Inequality



1.  **Positional measures:** they are the typical or representative values ​​of the variable under study.

**Arithmetic average:** The arithmetic mean formula calculates the mean or average of the numbers and is used to measure the central tendency of the data. It can also be defined as the sum of all the given observations to the total number of observations.

![](/images/arithmetic_mean.svg)

It is the most used measure. However, it is very sensitive to outliers and extreme observations that is why, if you have outliers in our data, we should use the median.

**Median:** For the calculation of the median, the data has to be arranged in ascending order, and then the middlemost data point represents the **median of the data**. The basic feature of the median in describing data compared to the mean (often simply described as the "average") is that it is not skewed by a small proportion of extremely large or small values, and therefore provides a better representation of a "typical" value.

![](https://www.gstatic.com/education/formulas2/472522532/en/median_formula.svg)

**Quantile:** quantiles are cut points dividing the range of a probability distribution into continuous intervals with equal probabilities, or dividing the observations in a sample in the same way. *Q*-quantiles are values that partition a finite set of values into *q* subsets of (nearly) equal sizes. There are *q* − 1 partitions of the *q*-quantiles, one for each integer *k* satisfying 0 < *k* < *q*.

Special cases of quantiles are **quartiles** when three values divide the data into four parts (Q1: leaves 25% of the observations to the left; Q2: median: leaves 50% of the observations to the left; Q3: leaves 75% of the observations to the left). **Deciles:** nine values ​​that divide the data into ten parts and **Percentiles** when 99 values ​ divide the data into one hundred parts

![](https://upload.wikimedia.org/wikipedia/commons/5/5e/Iqr_with_quantile.png)

Other positional measures:

**Geometric mean:** the geometric mean is a mean or average, which indicates the central tendency or typical value of a set of numbers by using the product of their values (as opposed to the arithmetic mean which uses their sum). For instance, the geometric mean of two numbers, say 2 and 8, is just the square root of their product, that is, ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/8a8d80ffa0ce79f7200d514844d4becd711f4f0b)

![](https://wikimedia.org/api/rest_v1/media/math/render/svg/026cae6801f672b9858d55935ec7397183dc3a36)

**Harmonic mean:** In mathematics, the harmonic mean is one of several kinds of average, and in particular, one of the Pythagorean means. It is sometimes appropriate for situations when the average rate is desired.

The harmonic mean can be expressed as the reciprocal of the arithmetic mean of the reciprocals of the given set of observations. As a simple example, the harmonic mean of 1, 4, and 4 is

![](https://wikimedia.org/api/rest_v1/media/math/render/svg/eca5906ddf61080e790c0d4df33f47e12da7d019)

The harmonic mean is one of the three Pythagorean means. For all *positive* data sets *containing at least one pair of nonequal values*, the harmonic mean is always the least of the three means, while the arithmetic mean is always the greatest of the three and the geometric mean is always in between.

**Robust means:** it is a trimmed mean by 10%. I mean, it is an arithmetic mean excluding the 10% of the largest values and the 10% of the smallest values)

**2. Dispersion measures:** the extent to which a distribution is stretched or squeezed.

**Variance:** variance is the expectation of the squared deviation of a random variable from its population mean or sample mean. Variance is a measure of dispersion, meaning it is a measure of how far a set of numbers is spread out from their average value.

**Standard deviation:** the standard deviation is a measure of the amount of variation or dispersion of a set of values. A low standard deviation indicates that the values tend to be close to the mean (also called the expected value) of the set, while a high standard deviation indicates that the values are spread out over a wider range. A useful property of the standard deviation is that, unlike the variance, it is expressed in the same unit as the data.

**Coefficient of variation:** the coefficient of variation (CV), also known as relative standard deviation (RSD), is often expressed as a percentage, and is defined as the ratio of the standard deviation ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/8d4ebf618b2ab59428af1b3f2e8e251c78d2f57a) to the mean ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/09bfcd47d25310c5187294b07d55e0954c2685f6) (or its absolute value, ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/f48e9c3009a01ad6bf5f22ce62c842ff99f19521)). The main advantage of this measure is that allows you to compare two variables that are in different units

![](https://wikimedia.org/api/rest_v1/media/math/render/svg/568abd256930542e24daafdd32985b401249c78d)

**Range:**  In statistics, the range of a set of data is the difference between the largest and smallest values, the result of subtracting the sample maximum and minimum. It is expressed in the same units as the data.

In descriptive statistics, the range is the size of the smallest interval which contains all the data and provides an indication of statistical dispersion. Since it only depends on two of the observations, it is most useful in representing the dispersion of small data sets.

**Interquartile range:** In descriptive statistics, the **interquartile range** tells you the spread of the middle half of your distribution.

Quartiles segment any distribution that’s ordered from low to high into four equal parts. The interquartile range (IQR) contains the second and third quartiles, or the middle half of your data set. The interquartile range is an especially useful measure of variability for skewed distributions.

![](https://cdn.scribbr.com/wp-content/uploads/2020/09/iqr_quartiles.png)

**Semi-interquartile range:** The semi-interquartile range is one-half the difference between the first and third quartiles. It is half the distance needed to cover half the scores.  The semi-interquartile range is affected very little by extreme scores.  This makes it a good measure of spread for skewed distributions.

**3. Shape measures:** Measures of shape describe the distribution (or pattern) of the data within a dataset. The distribution shape of quantitative data can be described as there is a logical order of the values, and the 'low' and 'high' end values on the x-axis of the histogram are able to be identified. they are the ones that measure the Kurtosis coefficient and Skewness. **Skewness tells you the amount and direction of skew** (departure from horizontal symmetry), and **kurtosis tells you how tall and sharp the central peak is**, relative to a standard bell curve.

**Skewness:** The **moment coefficient of skewness** of a data set is

skewness: *g*1 = *m*3 / *m*23/2

where

*m*3 = ∑(*x*−*x̅*)3 / *n*   and   

*m*2 = ∑(*x*−*x̅*)2 / *n*

*x̅* is the mean and *n* is the sample size, as usual. *m*3 is called the **third moment** of the data set. *m*2 is the **variance**, the square of the standard deviation.

If g1 = 0, the distribution is symmetric; if g1 < 0, the distribution is negatively skewed and if g1 > 0, the distribution is positively skewed.

**Kurtosis:** The other common measure of shape is called kurtosis. The **moment coefficient of kurtosis** of a data set is computed almost the same way as the coefficient of skewness: just change the exponent 3 to 4 in the formulas:

kurtosis: *a*4 = *m*4 / *m*22   and   excess kurtosis: *g*2 = *a*4−3

where

*m*4 = ∑(*x*−*x̅*)4 / *n*   and   *m*2 = ∑(*x*−*x̅*)2 / *n*

The excess kurtosis is generally used because the excess kurtosis of a normal distribution is 0. *x̅* is the mean and *n* is the sample size, as usual. *m*4 is called the **fourth moment** of the data set. *m*2 is the **variance**, the square of the standard deviation.

if g2 = 0 the ́distribution is Moscurtic; if g2 < 0, the distribution is platykurtic and if g2 > 0, the distribution is leptokurtic.

**4. Concentration measures:** Concentration measures study how the total values ​​of a variable are distributed. The most important concentration measures are the Lorenz curve and the Gini index. They are used in multiple fields like the study of economic inequality.

In economics, the Lorenz curve is a graphical representation of the distribution of income or of wealth. It was developed by Max O. Lorenz in 1905 for representing inequality of the wealth distribution.

The curve is a graph showing the proportion of overall income or wealth assumed by the bottom *x*% of the people, although this is not rigorously true for a finite population (see below). It is often used to represent income distribution, where it shows for the bottom *x*% of households, what percentage (*y*%) of the total income they have. The percentage of households is plotted on the *x*-axis, the percentage of income on the *y*-axis. It can also be used to show distribution of assets. In such use, many economists consider it to be a measure of social inequality.

![](https://upload.wikimedia.org/wikipedia/commons/thumb/5/59/Economics_Gini_coefficient2.svg/260px-Economics_Gini_coefficient2.svg.png)

The **Gini coefficient** measures the inequality among values of a frequency distribution, such as the levels of income. A Gini coefficient of 0 expresses *perfect equality*, where all values are the same, while a Gini coefficient of 1 (or 100%) expresses *maximal inequality* among values. For example, if everyone has the same income, the Gini coefficient will be 0. In contrast, if for a large number of people only one person has all the income or consumption and all others have none, the Gini coefficient will be nearly one.

![](https://upload.wikimedia.org/wikipedia/commons/thumb/c/cc/Map_of_countries_by_GINI_coefficient\_%281990\_to\_2020%29.svg/800px-Map_of_countries_by_GINI_coefficient\_%281990\_to\_2020%29.svg.png)

**Atkinson index:** The Atkinson index (also known as the Atkinson measure or Atkinson inequality measure) is a measure of income inequality developed by British economist Anthony Barnes Atkinson. The measure is useful in determining which end of the distribution contributed most to the observed inequality.  The index can be turned into a normative measure by imposing a coefficient to weight incomes. Greater weight can be placed on changes in a given portion of the income distribution by choosing ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/a30c89172e5b88edbd45d3e2772c7f5e562e5173), the level of "inequality aversion", appropriately. The Atkinson index becomes more sensitive to changes at the lower end of the income distribution as ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/a30c89172e5b88edbd45d3e2772c7f5e562e5173) increases. Conversely, as the level of inequality aversion falls (that is, as ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/a30c89172e5b88edbd45d3e2772c7f5e562e5173) approaches 0) the Atkinson becomes less sensitive to changes in the lower end of the distribution. The Atkinson index is for no value of ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/a30c89172e5b88edbd45d3e2772c7f5e562e5173) highly sensitive to top incomes because of the common restriction that ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/a30c89172e5b88edbd45d3e2772c7f5e562e5173) is nonnegative.

The Atkinson index is defined as:

![](https://wikimedia.org/api/rest_v1/media/math/render/svg/696cc5917249b769a2974a0d5c45c1b92cafbc3b)

where ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/67d30d30b6c2dbe4d6f150d699de040937ecc95f) is individual income (*i* = 1, 2, ..., *N*) and ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/9fd47b2a39f7a7856952afec1f1db72c67af6161) is the mean income.

Entropy:  The formula for general entropy for real values of ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/b79333175c8b3f0840bfb4ec41b8072c83ea88d3) is:

![](https://wikimedia.org/api/rest_v1/media/math/render/svg/844bcf9d016e032d7e03f0de29b7733e36f0b8a9)

where N is the number of cases (e.g., households or families), ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/67d30d30b6c2dbe4d6f150d699de040937ecc95f) is the income for case i and ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/b79333175c8b3f0840bfb4ec41b8072c83ea88d3) is a parameter that regulates the weight given to distances between incomes at different parts of the income distribution. For large ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/b79333175c8b3f0840bfb4ec41b8072c83ea88d3) the index is especially sensitive to the existence of large incomes, whereas for small ![](https://wikimedia.org/api/rest_v1/media/math/render/svg/b79333175c8b3f0840bfb4ec41b8072c83ea88d3) the index is especially sensitive to the existence of small incomes.

**Inequality measures based on quantiles:** statistical measures based on quantiles are frequently applied to the analysis of income distribution as they comprise many popular inequality and poverty indices and indicators. Simple dispersion ratios, defined as the ratios of the income of the richest quantile over that of the poorest quantile, usually utilize deciles and quintiles, but in principle, any quantile of income distribution can be used. These measures do not depend on changes in scale and allow comparison of low, middle, and high income groups.

References:

https://www.cuemath.com/

https://en.wikipedia.org/

https://www.scribbr.com/statistics/interquartile-range/#:~:text=In%20descriptive%20statistics%2C%20the%20interquartile,half%20of%20your%20data%20set.

https://brownmath.com/stat/shape.htm

https://www.researchgate.net/publication/325980614\_Properties_of_Selected_Inequality_Measures_Based_on_Quantiles_and_Their_Application_to_the_Analysis_of_Income_Distribution_in_Poland_by_Macroregion


How to CONDUCT A EDA univariate 🗂️

In data science, projects usually starts by an inquiry from business people that are sent to Data or IT department to begin with the development of the product that fits the requeriments. After data, the chief of staff talks with their team and communicate the demand to their data scientist. So later, is very common, mostly for junior data scientist to start the development of a funcy machine learning / deep learning model that magically solve the problem that business needs. Actually, this is a hugh mistake. Because in 99% of the cases, this rapid model just don't fit the requirement because it was build without answering the right questions. This is why the scientific method came in place to follow the process of objectively establishing facts through testing and experimentation. The basic process involves making an observation, forming a hypothesis, making a prediction, conducting an experiment and finally analyzing the results.

If we extrapolate the methodology to the life cycle of data science projects, we can establish the following stages:

*   Understand in detail the objectives of the project: this first stage includes meeting with the customers (managers and experts of the company) and discussing and knowing in detail the goals of the project.

*   Collect the available data and information: it can be provided by the client databases, APIs, files or through surveys, etc. In this stage, we look forward to arriving at a clear understanding of all information.

*   Establish the hypotheses to be tested, working models, possible theories, and methodologies to use.

*   Estimate and validate previously established models. Contrast the hypotheses formulated.

*   Obtain preliminary conclusions.

*   Discuss previous conclusions with the client and reconsider the objectives if necessary.

*   Final conclusions and decision-making.

## What is the CRISP-DM methodology?

![](/images/CRISP-DM_Process_Diagram.png)

The CRISP-DM method was developed by IBM (which includes IBM SPSS Statistics and other products) and consists of a set of stages, so that at each stage you can go back to the previous one and improve the process. The procedure takes into account not only the understanding and needs of the client and the business, but also the scientific part of the data analysis. The sequence of the phases is not rigid. Moving back and forth between different phases is always required. The outcome of each phase determines which phase, or particular task of a phase, has to be performed next. The arrows indicate the most important and frequent dependencies between phases.

> “The scientific method trains the brain to examine and observe before making and statement of fact”

The stages are:

*   Business Understanding: In the first step, you need to pay attention to identifying the problem. You should identify the metrics used to measure success over baseline, which is the same as doing nothing. Secondly, the identification of the type of problem we are facing. Thirdly, identify key people within your organization and outside that will help you on the data science project. Fourthly, get the specifications, requirements, priorities, and budgets of the project. You need to know how accurate the solution needs to be. In addition, you should analyze if you should be built internally versus using vendor solutions. If a vendor is a solution, do a vendor comparison and checked their benchmark.
*   Data Understanding: The data understanding phase starts with initial data collection and proceeds with activities that enable you to become familiar with the data, identify data quality problems, discover first insights into the data, and/or detect interesting subsets to form hypotheses regarding hidden information.
*   Data Preparation: The data preparation phase covers all activities needed to construct the final dataset \[data that will be fed into the modeling tool(s)] from the initial raw data. Data preparation tasks are likely to be performed multiple times and not in any prescribed order. Tasks include table, record, and attribute selection, as well as transformation and cleaning of data for modeling tools.
*   Data Modeling: In this phase, various modeling techniques are selected and applied, and their parameters are calibrated to optimal values. Typically, there are several techniques for the same data mining problem type. Some techniques have specific requirements on the form of data. Therefore, going back to the data preparation phase is often necessary.
*   Evaluation: At this stage in the project, you have built a model (or models) that appears to have high quality from a data analysis
    perspective. Before proceeding to final deployment of the model, it is important to thoroughly evaluate it and review
    the steps executed to create it, to be certain the model properly achieves the business objectives. A key objective is
    to determine if there is some important business issue that has not been sufficiently considered. At the end of this
    phase, a decision on the use of the data mining results should be reached.
*   Deployment: Creation of the model is generally not the end of the project. Even if the purpose of the model is to increase knowledge of the data, the knowledge gained will need to be organized and presented in a way that the customer can use it. It
    often involves applying “live” models within an organization’s decision-making processes—for example, real-time personalization of Web pages or repeated scoring of marketing databases. Depending on the requirements, the deployment phase can be as simple as generating a report or as complex as implementing a repeatable data mining process across the enterprise. In many cases, it is the customer, not the data analyst, who carries out the deployment steps. However, even if the analyst will carry out the deployment effort, it is important for the customer to understand up front what actions need to be carried out in order to actually make use of the created models.

![](/images/Screen%20Shot%202022-09-21%20at%2000.01.57.png)

I would like to mention that there are two other basic methodologies: SEMMA and KDD. The methodology called SEMMA was proposed by the SAS Institute, and consists of the stages of Sample, Explore, Modify, Model, and Assess. On the other hand, the KDD (Knowledge Discovery in Databases) methodology refers to the process of finding knowledge in data and emphasizes the high-level application of data mining methods.

References

\[1] Gavin Wright, Scientific Method, <https://www.techtarget.com/whatis/definition/scientific-method#:~:text=The%20scientific%20method%20is%20the,and%20finally%20analyzing%20the%20results.>

\[2] IBM, CRISP-DM v0.1


THE CRISP-DM METHODOLOGY ⌨️

In the era that we are living in, where we are reading posts, sharing interesting moments, and buying cloth on social media and webs, it is well known that the community is feeding a big amount of data into the databases of those companies.

What do I mean by data? When we talk about data, it means facts that are registered and collected for further treatment or analysis. Let's put this in context. Let's suppose that we are looking for some shirts and jeans in Zara. We just choose our favorite clothes for the next season so then, we need to register or log in on the web and then enter our data to continue with the further payment plus delivery. In the end, the companies send us a ticket that has all the information related to our operation. This invoice represents a summary of the data that was collected such as the clothes that were bought, the price, the quantity, the payment method, and so on. This is the real data that the company collects from us to deliver the purchase, and it is ready to be treated and analyzed in the company's systems.

Currently, we are living in a world that is full of data that we leave not only on retail webs or on social media but also in other places like traffic or facility cameras that collect our movements in order to understand and study them. In addition to this, insurance companies could collect telematics data based on devices connected to the car that learn our behavior while we are riding a car to offer personalized products or services. This data is the raw material, the new gold in this age, and is useful to be treated to fully understand customer behavior.

## About Data Science

> “Data Science combines the techniques and methodologies of statistics, mathematics and computer science, with the aim of understanding, interpreting and inferring conclusions from data and making decisions.”

Now that we know what is data and the potential benefits that it has, we will continue to explain the definition of data science. One of the definitions that I learned during my master's degree was that data science deals with the study of data through the scientific method.  Data science combines the techniques and methodologies of statistics, mathematics, and computer science, with the aim of understanding, interpreting, and inferring conclusions from data and making decisions.  

Based on the definition, we can infer that data science is a multidisciplinary field where domain knowledge, statistics, computer science, and communication skills are used in combination. Furthermore, it is common to listen that data scientists are better statisticians than most programmers and better programmers than most statisticians. Continuing with the definition, we can see that it has different goals like collecting and understanding the data which is called descriptive analytics that answer the question of “*What happened?*”. Another goal is interpreting the data that is part of diagnostic analytics that answer the question of “*Why did this happen?*”. In this kind of analytics, we take the analysis a step further, because we are comparing coexisting trends or movements, uncovering correlations between variables, and determining causal relationships where possible. Finally, it talks about inferring conclusions that we can relate with predictive analytics that is used to make predictions about future trends or events and answers the question, *“What might happen in the future?”*

Without data, people make decisions based on biases and beliefs. So, in order to create value and take decisions with data fundamentals, companies use their data, explore it, analyze and predict future behaviors with them. In this scenario, data scientist takes place in order to extract the data, understand it and apply different techniques to solve problems. But, how do they deliver value to the organization? It is by the following goals:

*   Revenue increase: data science can be applied to find and refine a target customer base to generate more revenue. In sales, specifically, lead management, models can analyze past customers and score leads, resulting in greater sales efficiency. Not only past in-house customer data is used, but also social media interaction for scoring. Companies that applied artificial intelligence lead scoring increased by 50% more appointments and a 60% reduction in call time *\[2]*. Better leads close sales deals faster, which creates more revenue.

*   Cost reduction: In my own personal experience in financial services, we have applied data science concepts to reduce customer turnover and fraud risk. Using predictive models, we predict when customers are likely to get into trouble and offer assistance to prevent breaches of contract. As for fraud prevention, we analyzed client data to detect anomalies that might be fraudulent, which leads to a potential savings of millions per year.

*   More efficient processes: for example data science applied with real-time reporting *\[4]*, customer interactions are more effective by allowing service representatives to better understand consumers while interacting with them. Think about how many times you’ve made a call to a customer service line and waited on hold only to answer a bunch of questions, be put on hold again, and then have to answer all the same questions with everyone else you talk to. With real-time reporting, customer service agents have instant access to the questions callers have already answered, shortening the call and hold times while improving the overall experience.

*   Knowledge discovery: it is related to the extraction of knowledge of the data collected that was previously unknown, and it has useful information

*   More efficient decision-making: With the help of data science, decision-makers are now able to make better-informed choices than in the past by shaping and filtering the data their organizations have collected. By using this data, they are able to formulate predictions of the future based on what would happen if they decided to take their organization in a brand-new direction, for example, or how they would rebuild themselves after a financial disaster, among a variety of other things. For example, in Airbnb, the marketing department thought one of their campaigns was off due to bad metrics *\[3]*. A data scientist found the root cause to be a group in Asia using their site to browse neighborhoods. As a result, the marketing department redirected that group to a travel site, which clarified their marketing metrics and saved resources instead of launching an inefficient marketing initiative.

References

\[1] Lester Leong, How Data Science makes money for businesses (2019), <https://towardsdatascience.com/how-data-science-makes-money-for-businesses-203daba010c1>

\[2] T. Capone, How Data Science Can Help Your Enterprise Generate More Revenue (2018), <https://conceptainc.com/blog/how-data-science-can-help-your-enterprise-generate-more-revenue/>

\[3] N. Patel, How Airbnb Uses Data Science to Improve Their Product and Marketing (2019), <https://neilpatel.com/blog/how-airbnb-uses-data-science/>

\[4] Landon Starr, How Data Science Can Improve Business Efficiency (2018), <https://blogs.oracle.com/ai-and-datascience/post/how-data-science-can-improve-business-efficiency>


Blog

Git Integration with Power BI

The magic of choosing an EFFECTIVE visual

Shiny Web App to visualize key financial metrics

How to CONDUCT A EDA univariate 🗂️

THE CRISP-DM METHODOLOGY ⌨️

What is DATA SCIENCE? 🚀