Business Science

Two months ago, we launched a revolutionary new way to teach data science for business: #BusinessScienceUniversity. Our first course, Data Science For Busines (DS4B 201) teaches our methodologies and cutting-edge tools including #H2O on a real-world churn problem. This is the way #DataScience should be taught in businesses. Learn more at https://lnkd.in/gnkXeqK #DataScience AND #MachineLearning #Education FOR #Business

A new trend is emerging: Rise of the Generative AI Data Scientist This is what's happening (and how to prepare): After spending 12 months learning Generative AI, I'm 100% certain that Data Science teams will change forever. But 99% of Data Scientists are unprepared for Generative AI. Their loss can be your gain. 1. This is what's happening. I love data science. I've seen how much money Data Science can save companies. I've personally saved companies over $25,000,000 with data science. But, AI has changed the Data Science landscape. 2. With generative AI we can create AI Agents. These AI agents operate like real people with roles, skills, and domain knowledge. AI Agents can do things like: - Forecast Time Series - Write SQL - Analyze Data - Make Machine Learning Models - Execute Python Code - Create Reports - Send Emails 3. That's why a new role is rising: The Generative AI Data Scientist Here's how to prepare: On Tuesday, November 26th, I'm hosting a live Python workshop: AI for Data Scientists You and I will go through one of my BEST AI projects: A Time Series Forecasting AI Agent. Register here: https://lnkd.in/eGUn-rzd

My favorite R package for ultra-fast exploratory analysis: The R package is called correlationfunnel. (Yes, I built it.) What correlationfunnel does is: * Speeds Up Exploratory Data Analysis * Improves Feature Selection * Gets You To Business Insights Faster Learn more here: https://lnkd.in/eVeEFyXq Want to become a data scientist with R? I've created a complete system called the 5-Course R-Track that teaches you everything you need to become a Business Scientist with R. 1,848 Lessons, 152.9 Hours of Video, and 24 Challenges to Test Your Skills Learn more in my free R masterclass: https://lnkd.in/eejE6yiR

Made with R: About the project (and get the code here): https://lnkd.in/dKReMy4 Want to become a data scientist with R? I've created a complete system called the 5-Course R-Track that teaches you everything you need to become a Business Scientist with R. 1,848 Lessons, 152.9 Hours of Video, and 24 Challenges to Test Your Skills Learn more in my free R masterclass: https://lnkd.in/eejE6yiR

The Number 1 Skill that helped me grow my data science career? Time Series Forecasting. Here's 2 years of help in 2 minutes: 1. Why should you learn forecasting? Companies need better forecasting. It solves a multi-million dollar problem for them. Most data scientists don't know how to do forecasting. Competition is low. By picking up this skill, you can get ahead of 80% of data scientists. 2. How to Learn Time Series? If you are an R user, I've built an full system of packages: 1. Modeltime 2. Timetk 3. Modeltime.ensemble 4. Modeltime.resample 5. Modeltime.h2o 6. Modeltime.gluonts I'll focus on the 1st 2 packages to help you get started. 3. Modeltime Use this R library to forecast and compare: - Classical Models (ARIMA, ETS) - Newer Methods (Prophet) - Machine Learning (Tidymodels) GitHub: https://lnkd.in/dtGuZcy 4. Timetk Use this R library to perform time series analysis: - TS Data Aggregations & Manipulations - TS Visualization (Seasonality, Autocorrelation) GitHub: https://lnkd.in/dGSKpSk With those 2 R packages you can get ahead of 80% of data scientists. If you want to go further, I have a High-Performance Time Series course with 45.8 hours of video training: https://lnkd.in/evQKZzhC If you'd like to hear more about the blueprint to prepare yourself for a 6-figure data science for business career with R... ... then I'd like to help. I've created a complete system called the 5-Course R-Track that teaches you everything you need to become a Business Scientist. 1,848 Lessons, 152.9 Hours of Video, and 24 Challenges to Test Your Skills Learn more in my free R masterclass: https://lnkd.in/eejE6yiR

The best beginner book on time series? FPP: Forecasting Principles and Practice Let's dive in: Some may consider FPP the bible of time series. I agree. Start with FPP Version 2 or 3, and you won't go wrong. This is what I like: 1. Time Series Visualizations There's no better intuition for time series than a visualization. - Time Plots - Seasonal Plots - ACF Plots All are absolutely critical to understanding time series patterns. 2. Decomposition Many methods are covered. Arguably the most important is STL. Stands for Seasonal-Trend-Loess. I use STL for detecting outliers and discovering trends. 3. Forecasting with Exponential Smoothing ETS models need to be in every forecaster's toolbox. These are state-space models that are simple and fast to run. 4. ARIMA (and variants) ARIMA is another foundational time series forecast model that every forecaster needs to know. -ARIMA -SARIMA -SARIMAX All are extremely beneficial for forecasters. 5. Hierarchical Models Forecast reconciliation and distributional reconciliation are important for many companies that have hierarchical structures within their demand data. Methods covered: - Top-Down - Bottom-Up If you'd like to hear more about the blueprint to prepare yourself for a 6-figure data science for business career... ... then I'd like to help. I've created a complete system called the 5-Course R-Track that teaches you everything you need to become a Business Scientist. 1,848 Lessons, 152.9 Hours of Video, and 24 Challenges to Test Your Skills Learn more in my free R masterclass: https://lnkd.in/eejE6yiR

Can Generative AI Agents make time series forecast models? This is what I found out: 1. Generative AI is the future of Data Science Over the last 12 months I've been learning Generative AI. And I am 100% certain that GenAI is the future. AI Agents can: - Write SQL code - Make Machine Learning Models - Execute Python Code - Write Reports - Send Emails And basically, every task I do. 2. Example: Building A Time Series Forecasting Agent To test my theory, I set out on a project to replicate my time series forecasting process with AI. I built AI agents to: - Extract and aggregate time series data from a SQL Database - Make a forecast model using XGBoost - Chart the forecast - And a supervisor agent that decides when to get each agent involved 3. The results were shocking The AI Forecasting Team completed my forecasting process in under 1 minute. It handled tons of different forecasting requests from easy to difficult. 4. Want to learn this project? On Tuesday, November 26th I'm hosting a live AI For Data Scientist Workshop where I'll have my Time Series Forecasting AI ready to share. Register here (500 seats): https://lnkd.in/eGUn-rzd

Microsoft Forecasting Framework for Finance Time Series 🚀 An R package that extends Modeltime for automated forecasting for financial forecast =) The finnts is an R package from Microsoft for automated forecasting for financial forecast. The framework includes the following features: ✅ Automated feature engineering and feature selection ✅ Training models with backtesting, scoring, and ranking ✅ Supports more than 25 models for univariate and multivariate time series data ✅ Supports time series with daily, weekly, monthly, quarterly, and yearly frequency ✅ Azure integration Learn more here: https://lnkd.in/e4Q8rjMk Want to become a data scientist with R? I've created a complete system called the 5-Course R-Track that teaches you everything you need to become a Business Scientist with R. 1,848 Lessons, 152.9 Hours of Video, and 24 Challenges to Test Your Skills Learn more in my free R masterclass: https://lnkd.in/eejE6yiR

A New Role is Emerging: The Generative AI Data Scientist Let me explain what's coming (and how you can take advantage): Companies everywhere are hiring for a new role: GenAI Data Scientist. Salaries range from $115,000 to $264,200 a year. But what makes these people so special (and why are 99% of data scientists unqualified)? 1. 99% of data scientists are not building the GenAI skills companies actually want Most data scientists think ChatGPT is GenAI. (It's not) GenAI that companies are seeking is using machines to automate processes. 2. What do companies actually want from a GenAI Data Scientist? Companies need to reduce costs and increase revenue. They do this through automating processes and cutting the fat. 3. Going from 10 data scientists to 1 GenAI Data Scientist When a company like Apple, NVIDIA, Goldman Sachs, and SAP hires a GenAI Data Scientist... ...They can reduce large 10+ person data science teams by automating the routine analytics. This is a massive cost savings. 4. That's Why 1 GenAI Data Scientist Can Earn $200,000+ Because 1 GenAI Data Scientist actually a cost savings vs $1,500,000 data science team. This is how to prepare for this trend: 5. How to prepare On Tuesday November 26th, I'm hosting a live workshop on how to build and deploy Generative AI Data Science solutions. I'll share one of my top AI projects: Time Series Forecasting Agent. Register here (500 seats): https://lnkd.in/eGUn-rzd

R is insane for time series (and fixing messy data). Introducing imputeTS: 1. What is imputeTS? The imputeTS package specializes in (univariate) time series imputation easier. It offers several different imputation algorithms, plotting tools, and summary printing utilities. 2. Visualizing Missing Values in Time Series It's easy. Just use ggplot_na_distribution() 3. Imputation Algorithms imputeTS has 10 different interpolation functions: 4. Imputation Plotting Functions To help understand missing values, imputeTS comes with 4 plotting functions: Get imputeTS on GitHub here: https://lnkd.in/eUkWe_mx Want to become a data scientist with R? I've created a complete system called the 5-Course R-Track that teaches you everything you need to become a Business Scientist with R. 1,848 Lessons, 152.9 Hours of Video, and 24 Challenges to Test Your Skills Learn more in my free R masterclass: https://lnkd.in/eejE6yiR