Restaurant-Tips

Abstract

The project was prepared and submitted by C. Pinheiro, E. Carvalho, M. Nazarkovsky, G. Piovesan, I. Barros within the Brazilian "Bootcamp Data Science na prática" by Neuron.

See also here

The certificate

The project file is uploaded in this repository as a PDF

The dataset was provided by the Bootcamp administration as a tips.xlsx from Kaggle

Data types

Continious data: Total bill, Size of group, Tip Nominal (categorical) data: Sex, Smoker (Smoker/Non-smoker), Week's day, Time (Lunch/Dinner)

Objectives

The project is devoted to the analysis of factors (regressors) which may affect the tips. As a result, the objective is to determine the conditions for the restaurant for a promotion and/or happy hours

Methods and Approaches

● Means Comparison (Tukey-Kramer) ● ANOVA ● Welch’s Method ● Equality of Variances tests.

Results and Discussions

1. No statistically significant difference has been found in Tip variances by the following factors: Sex, Smoker, Time e Days. But some observations for Time and Days are claimed;
2. There is a positive correlation between Tips, Total_bill and Size of the Group;
3. The biggest contribution to Tips and Total bill is made by group of 2 clients in the dinner time;
4. In terms of the week's days, the biggest contribution to Tips and Total bill is made by a groups of 2 clients, whose percentage is lowered on Sat and Sun;
5. Since Tips and Size of the Group are the lowest on Fri, it would be reasonable to make promotions this day for the groups of 5 or more persons (colleagues, classmates friends, happy hour, birthday parties).
6. More specifically: the promotions for the lunch time on Fri with the focus on the groups of 5 or more persons will improve the run of the restarant's business