The Home Improvement Company (HIC) operates five stores in a large regional area. The company wished to study the relationship between x, house value ($'000) and Y, yearly expenditure on house maintenance ($ upkeep) as a basis for promoting the types of products it sells.
A random sample of the owners of 40 houses was taken. They were asked to estimate their maintenance expenditure on the types of products HIC sells. HIC then contacted real estate agents to value the houses. See the data file.
The task of modelling the relationship between Upkeep (Y) and Value (x) was assigned to an analyst within HIC. All the data had been validated and the analyst was instructed not to delete data for 'statistical convenience'.
Part (a)
You are to decide which of the three models below are most appropriate, where Y = estimated maintenance spend (upkeep) and x = house value
Y= β_{0} + β_{1}x +ε
Log(Y) = β_{0} +β_{1}x +ε
√y = β_{0} +β_{1}x +ε
Your report for Part (a) should start with an exploratory analysis of the data, and then three sections, detailing the analyses of models (1), (2) and (3) and then a short recommendation.
You should analyse and report leverage points and outliers where appropriate using the template below but you cannot remove any data.