Data Analytics - Blog Posts

Nothing aggravates me more than the ‘Ask app not to track’ notification, because WHAT DO YOU MEAN I HAVE TO ASK??? I’m not knocking gently on the door, pushing it open, and saying ‘ummmm… can you maybe not take all of my data and analytics and make me feel like less of a person and more of an opportunity for capitalism?…. no…? Okay, sorry for bothering you.’ NO! I want them to NOT TRACK ME.

The GOAL is not to learn Power BI, the GOAL is to BECOME a Data Analyst.

But Power BI kicking my ass. Tf I need this for? 😐

Girl u are stubborn 🤣

Live AQ map

Your Essential Guide to Python Libraries for Data Analysis

Here’s an essential guide to some of the most popular Python libraries for data analysis:

 1. Pandas

- Overview: A powerful library for data manipulation and analysis, offering data structures like Series and DataFrames.

- Key Features:

  - Easy handling of missing data

  - Flexible reshaping and pivoting of datasets

  - Label-based slicing, indexing, and subsetting of large datasets

  - Support for reading and writing data in various formats (CSV, Excel, SQL, etc.)

 2. NumPy

- Overview: The foundational package for numerical computing in Python. It provides support for large multi-dimensional arrays and matrices.

- Key Features:

  - Powerful n-dimensional array object

  - Broadcasting functions to perform operations on arrays of different shapes

  - Comprehensive mathematical functions for array operations

 3. Matplotlib

- Overview: A plotting library for creating static, animated, and interactive visualizations in Python.

- Key Features:

  - Extensive range of plots (line, bar, scatter, histogram, etc.)

  - Customization options for fonts, colors, and styles

  - Integration with Jupyter notebooks for inline plotting

 4. Seaborn

- Overview: Built on top of Matplotlib, Seaborn provides a high-level interface for drawing attractive statistical graphics.

- Key Features:

  - Simplified syntax for complex visualizations

  - Beautiful default themes for visualizations

  - Support for statistical functions and data exploration

 5. SciPy

- Overview: A library that builds on NumPy and provides a collection of algorithms and high-level commands for mathematical and scientific computing.

- Key Features:

  - Modules for optimization, integration, interpolation, eigenvalue problems, and more

  - Tools for working with linear algebra, Fourier transforms, and signal processing

 6. Scikit-learn

- Overview: A machine learning library that provides simple and efficient tools for data mining and data analysis.

- Key Features:

  - Easy-to-use interface for various algorithms (classification, regression, clustering)

  - Support for model evaluation and selection

  - Preprocessing tools for transforming data

 7. Statsmodels

- Overview: A library that provides classes and functions for estimating and interpreting statistical models.

- Key Features:

  - Support for linear regression, logistic regression, time series analysis, and more

  - Tools for statistical tests and hypothesis testing

  - Comprehensive output for model diagnostics

 8. Dask

- Overview: A flexible parallel computing library for analytics that enables larger-than-memory computing.

- Key Features:

  - Parallel computation across multiple cores or distributed systems

  - Integrates seamlessly with Pandas and NumPy

  - Lazy evaluation for optimized performance

 9. Vaex

- Overview: A library designed for out-of-core DataFrames that allows you to work with large datasets (billions of rows) efficiently.

- Key Features:

  - Fast exploration of big data without loading it into memory

  - Support for filtering, aggregating, and joining large datasets

 10. PySpark

- Overview: The Python API for Apache Spark, allowing you to leverage the capabilities of distributed computing for big data processing.

- Key Features:

  - Fast processing of large datasets

  - Built-in support for SQL, streaming data, and machine learning

 Conclusion

These libraries form a robust ecosystem for data analysis in Python. Depending on your specific needs—be it data manipulation, statistical analysis, or visualization—you can choose the right combination of libraries to effectively analyze and visualize your data. As you explore these libraries, practice with real datasets to reinforce your understanding and improve your data analysis skills!

https://www.bloomberg.com/graphics/2024-opinion-biden-harris-accomplishment-data/

"In the race between Vice President Kamala Harris and former President Donald Trump, the polls are tight and their policy plans are underwhelming. But there is another way to compare their ability to do the job that gets far too little attention: Both candidates have a record to run on — and the data tell their own story."

Even though it's past election day, it's still a good thing to inform people what has been happening. Even as I talked with some folks the day after, I realized pretty quickly that there's a lot of people who don't know what's going on in the country when it comes to the topics that people like to blow up about. Also, DO NOT FORGET that the president is the figurehead, it's all the people they employ that do all the work. If you voted at all, you were voting for a staff of people, not just one person.

I just want to highlight a few graphs.

Btwn year 2-3 of Trump, it was already increasing, that's right before COVID became a thing. Trump's approach to life is emboldening violence.

Again, Trump didn't do anything spectacular besides being on par with most other presidencies, until the administration truly failed. Biden's administration blew the past 3 presidents out of the water. And yet people are still crying over the economy.

Speaking of economy, the article highlights there were proposals to try to reign in price gouging. Some folks might be familiar with the concept of rent control and the like. It was disappointing that these things were not enabled. But for conservatives, it's something they don't want to hear because they don't want to let the government manage business but cry out why the government is not managing business.

I'm just going to copy/paste what the article has for this one:

Presidents love to take credit for a booming market and disavow a sagging one. The truth is they have little to do with either.

Still, some policies are better for asset prices than others, even if it’s difficult to quantify how much. Trump wants to extend the 2017 tax cuts and lower corporate taxes, which could boost spending and corporate earnings, pushing stocks higher. On the flipside, his proposed tariffs could hurt companies, particularly if US multinationals face retaliatory levies.

Harris wants to raise corporate taxes and steer that money to workers and families as tax credits and support for first-time homebuyers. Companies would presumably be worse off from higher taxes, although putting money in consumers’ pockets is generally good for business. The risk for fixed income is clearer: Both Harris and Trump’s proposals are likely to result in higher deficits and debt, which could stoke inflation and force the Fed to raise rates, giving bonds a whacking.

Most likely, though, what drives markets will have little to do with these plans.

The day after the election, of course I heard from the mouths that don't care that the stock was booming. That's because stocks are speculation and really just operate off perception. The people who are in dire need of funds don't have stocks. Those are the people who will get no benefit from increased stocks and yet they see it and think it's great when it only serves a few.

Don't forget, taxes fund the functioning of government. It's absolutely needed. Do you want taxes to come from you or from big corporations?

Honestly, no administration is perfect. But after this abysmal election, I'm not going to shut up about being informed of what's really going on. The truth of how this country is performing and making sure that people are asking the right questions. I already have some questions out of this report, such as what's going on at the state level? The federal level can only do so much. As an example, taxes at the state and city levels can help mitigate and perform better for their own community.

And if we work to better our own communities, maybe that can help funnel up.

Banks are in the pursuit of finding feasible ways of securing systems and services from rampant cybersecurity threats. There have been enough financial frauds in the recent past to drive banking institutions into considering novel methods of risk monitoring and prevention. Data analytics offers impactful solutions in this respect. 

Analytics solutions that have been customized to deliver anti-fraud services to banking firms are enabling a modern, technology-driven approach in managing security concerns. Intelligence and automation are the mainstays of advanced analytics solutions, making fraud analytics the key for relevant and reliable security capabilities.

The extent to which fraud analytics can save banks from exposure to risks depends on the security features provided by the vendors, as well as on a banking firm's ability to implement the solutions optimally. This article talks about the events leading up to the widespread adoption of fraud analytics in banking and what banks can expect from the analytics tools.

Scope of Overcoming Challenges

Adopting advanced technology has become viable from a financial and utility point of view. High network speeds, cloud computing, artificial intelligence, machine learning, IoT, and data analytics are making their way into every-day business applications. As advantageous as these technologies might be for banks, they have given rise to cybersecurity risks. Cybercriminals are....

The trend of making services customer-oriented is a direct result of rising demands among customers for better services, and the best way for banks to achieve it is through the use of analytics. Technology is the most crucial aspect of the digital age, but so are the customers and their changing expectations on service providers. Improved customer services enhance business opportunities and increase brand value, due to which banks are prioritizing it.  

Under the traditional approach, banks collected customer information to build databases that could be used to disburse services. Now, customer data has become easily accessible, and analytics solutions are empowering banks to leverage the data for improving customer services.

The use of analytics in the banking sector is not new. Several processes are already using data analytics solutions to hike the efficiency of operations across departments. However, the potential of analytics in improving customer service is incredibly high, making it an indispensable tool for banks. The following list throws light on the factors which make analytics important for CIOs in their quest to help banks improve customer service.

• Handling Vast Data Volumes Effectively

There are several constraints when it comes to handling analytics through traditional approaches. With the exponential rise in the amount of customer data available with banks, manual analytics is out of the question now. Advanced data analytics solutions are equipped...

Big data analysis is the buzzword in the technology domain has gathered massive traction in recent time. As the customer touch points are increasing from digital communications like social media and emergence of new innovative technologies like IoT, companies now have data sources that give real-time information. 

It is forcing enterprises to rely on it for more profound insights which are helping organizations to grow. Here are five significant data analytics trends that will be the talk of the technology world in 2019 and beyond.

https://goo.gl/XEcJHK

How the Importance of DevOps for Data Science is Growing

DevOps consists of infrastructure provisioning, continuous integration, and deployment, configuration management, monitoring, and testing.  DevOps teams and development teams have been working closely to manage the lifecycle of applications effectively.

More on this here: https://goo.gl/vy5E81

Streamlining Smart Cities with Geographic Information System

Decades of unplanned urbanization has induced rampant pollution, sanitation hazards, inadequate housing, slums, and transport perils, Along with that, extreme weather patterns have become a global concern. It is about time data-driven adaptive urbanism—smart cities—gains momentum, building cities as models of a global change.

This model will pave the way for connected spatial data flow, enabling every resident in the city to comprehend the goals the community needs to accomplish. The residents can utilize applications to keep themselves informed about recreation, business, property, transportation, taxation, and more.

These smart cities function round-the-clock in a cloud environment and are entirely based on knowledge and innovation for best and sustainable civilian services. The definitions for smart cities vary depending on geographies, deployment, and scaling. However, in a data-driven environment Read More 

Interesting Read: At the Pinnacle of Smart City Aspirations

The Stepping Stones for Decision Analysis and Optimization

While data analytics do help garner deep insights for smarter decision-making, the insights required varies based on the technology and the analysis approach and procedures used.

 Businesses need to ensure that they have a business intelligence architecture or data warehouse that offers a convenient and multi-faceted analytical ecosystem optimized for effective analysis of diverse datasets.

Descriptive Analytics

Descriptive analysis is all about using the past performance, understanding its nature by mining the historical data to analyze the reason for a previously occurred success or failure. Descriptive models enable businesses to classify the prospects or parameters by consolidating relationships in data.

Diagnostic Analytics

An advanced level of analytics, diagnostic analysis, dissects the data to answer the reason behind a specific event. Methods used to characterize the data include data discovery, mining, drill down, and  Read More

Source: APAC Business Intelligence