Sukhpreet1987
commented
6 years ago Hypothesis testing was carried out in R programming and derived conclusions as below:
a. Hourly Trends- I can say that 7-9 and 17-19 hours have high bike demand; 10-16 hours have average bike demand 0-6 and 20-24 hours have low bike demand. b. Daily Trend - distribution of casual and registered users in which outliers are removed from data using logarithmic transformations.Registered users follow same trend as count than casual users where Hour is a significant factor. c. Rain- Weather has 4 types, depicts less bike use on heavy rainy days(4 - Heavy Rain) d. Time- Concluded that 2012 has higher bike demand. d. Temperature- are continuous variables, considered correlation analysis for testing hypothesis and concluded temp is positively related with demand of bikes Pollution and traffic - No data present hence no prediction for hypothesis test is carried out ---## R code uploaded - Hypothesis Testing.R---
Week 2
Hypothesis generation and testing
For this study different bike demand hypothesis were influencing factors which are as given below: a. Hourly Trends: More demand in office travelling hours and less demand in night time from 10pm to 4am as well different pattern in morning and evening bike usage. b. Daily Trend: casual users use less bikes and registered users use bike mostly on weekdays than weekend or holiday c. Rain: Rainy day as well as humid weather cause less bike demand than sunny days d. Temperature: Checking whether there is positive correlation in the temperature based on data e. Pollution: Are pollution levels affecting bike demand f. Time: Registered users have better time contribution than casual users. g. Traffic: Higher traffic means more bike demand than taxi and personal vehicles