Notebooks and descriptions • Contact Information
| Notebook | Brief Description |
|---|---|
| retail-store-expansion-analysis-with-lasso-and-ridge-regressions | Based on a dataset containing the spirits purchase information of Iowa Class E liquor licensees by product and date of purchase this project provides recommendations on where to open new stores in the state of Iowa. To devise an expansion strategy, I first needed to understand the data and for that I conducted a thorough exploratory data analysis (EDA). With the data in hand I built multivariate regression models of total sales by county, using both Lasso and Ridge regularization, and based on these models, I made recommendations about new locations. |
| conjoint-analysis | Conjoint analysis is a technique that allows researchers to predict consumers' choice share. The analysis can be programmed using standard question types, such as the MaxDiff variation of the Matrix Table question. Instead of directly asking the survey respondents which attributes they find most relevant, conjoint analysis asks respondents to evaluate potential product profiles which include multiple product features |
There are several ways to show to respondents the product profiles. In Choice-Based Conjoint (CBC) respondents are shown multiple product conceptsn and asked which option they would choose. By varying the features shown to the respondents and observing their responses to the product profiles, one can statistically deduce the most desired product features and which attributes have the most impact on choice. The end result is a set of preference scores or part-worth utilities for each level of each attribute. In this notebook I show how to use Python to calculate the utilities. The notebook is heavily based on this course and this book.|
Feel free to contact me:
- Email: [email protected]
- GitHub: marcotav
- LinkedIn: marco-tavora
- Website: marcotavora.me
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent