Evidence (Bloque 1), Stats for AI
1 Objective
You have historical monthly data of all milk products sold in the channel of price club stores at the national level.
You are supposed to be a data engineer and analyst in one of the 3 big manufacturers of milk products sold in this channel.
To know which Manufacturer you work for, take your matricula, add the 8 digits and take the mod 3 plus 1. For example, if your matricula is A01234567, then adding 1 to 7 you get 28, then 28 mod 3 gives you 1, then add 1 and you get 2, which is the FAB #2 (fab2).
The objectives of your analysis are the following:
Provide understanding of the milk category by manufacturer (fab) in terms of a) market share composition by year, and b) sales and price performance over time. Identify the top fabs that make 80% of the market share for the last 12 months. Show relevant descriptive statistics according to your analysis and assumptions.
Provide initial correlation analysis to understand important relationship within the data. You have to justify how and why you did this correlation analysis. Explain your results
For 1) and 2) Present your analysis in tables and selected graphs to provide a good understanding of the data.
- Design and run regression model(s) to estimate price sensitivity (direct price elasticity) for the top product of your manufacturer (fab#). Identify the top product as the product that sold the most for the last 12 months. For this model, only consider the price change as explanatory variable (do not include other factors such as seasonality or tren).
You have to present the regression output for the model and provide a clear interpretation for the regression output: parameter estimates (beta coefficients), t-Statistics, pvalues, 95% confidence intervals, and R-square.
Propose how you could improve your model design to incorporate the following factors: cross-price elasticity with other products, seasonality and growth trend. Here you have to explain your model design, not run the model.
For your model (of the top product), add variables to include the effects of seasonality and growing volume trend.
Run the model, show the regression output and interpret the outputs
In addition, respond to the following questions:
Did the estimates of direct price elasticity change compared with the previous model? What happened with the statistical significance of the price elasticity? Why do you think the estimates change, and which estimate is the best? Explain clearly with your own words.
- Propose how you could forecast price and sales of the top product for the next 24 months. Which types of models you can use to do this. You have to do your own research. Consider not only regression models, but any type of machine learning/ AI models.
2 PART 1 - Data wrangling and data understanding
You have to download (from Canvas) the following files:
- DATOS_VENTAS.xlsx: historical monthly data for all products sold for the last 3-4 years
You have to:
- Do a correct data wrangling/cleanning to prepare the data for descriptive statistics and modelling:
You have to check for null values, possible duplicate records (1 same product with 2 product ID, or 1 product ID with 2 different product names),
Then you have to do the right data collapsing, possible reshaping, and variable transformations, and come up with the right data structure for analysis. Remember dataset structure types (long-format, wide-format).
Do the corresponding data analysis to respond to the objective 1)
Run and interpret the regression model for your fab
3 PART 2 - Data modelling
You have to design, run models to respond to the objective and questions stated in 3), 4) and 5)
Go to Canvas and check DEADLINES for each part