Use any techniques (correlation, regression, machine learning, deep learning, etc.) to answer the following two questions: ?

Use any techniques (correlation, regression, machine learning, deep learning, etc.) to answer the following two questions:

·  What factors significantly impact the amount of wines (AmountWines)?

o  Due date: Monday 12/13/2021 11:59pm CT

o  Data: Canvas-assignment-Final Project

o  Submit your codes via Canvas-assignment-Final Project

o  Submission format: r.script and Single Excel summary output

You must follow below steps. Otherwise, your code won’t work on my machine

§  Manually set your working directory (from toolbar – Session), DO NOT hardcode the file path.

§  Only use the following code to read data from your local drive

data_independent  <- read.csv(file = 'data – for student – independent variables.csv')

data_dependent <- read.csv(file = 'data – for student – dependent variable.csv')

y <- data_dependent$AmountWines

§  Enter your name, such as:

first_name<-'Yan'

last_name<-'Lang'

§  Assign “new_data”, lower-case, to represent all independent variables you decided to use

§  Assign “model”, to represent your model’s name

§  Assign “preds”, lower-case, to represent your y-predictions based on the results of your model. If your data has 1500 rows, then you should see 1500 rows of predictions.

§  write below code after calculating your predictions

rmse<- sqrt(mean((y – preds)^2))

§  Combine, “first_name”, “last_name”, and “rmse” as one single csv output file, name the csv file as: finaloutput

§  At the end of your scripts, write the following code (this is for TA, when testing your code, don’t run those lines.)

source("predictdata.R")

·  What are your business suggestions/recommendations to the CEO?

o  Due date: Monday 12/13/2021 11:59pm CT

o  Submit your file via Canvas-assignment-Final Project

o  Submission format: Single Word file

§  Data visualization

§  Executive summary reports

§  etc.