# Linear Regression predicting price of wine

# Read in data
wine = read.csv("wine.csv")

## Warning: cannot open file 'wine.csv': No such file or directory

## Error: cannot open the connection

str(wine)

## Error: object 'wine' not found

summary(wine)

## Error: object 'wine' not found

# examine correlations

cor(wine)

## Error: object 'wine' not found

round(cor(wine), 2)

## Error: object 'wine' not found

wineReordered = wine[c("Price", "Year", "WinterRain", "AGST", "HarvestRain", 
    "Age", "FrancePop")]

## Error: object 'wine' not found

round(cor(wineReordered), 2)

## Error: object 'wineReordered' not found

pairs(wine)

## Error: object 'wine' not found

## Correlogram example
install.packages("corrgram")

## Installing package into '/Applications/RStudio.app/Contents/Resources/R/library'
## (as 'lib' is unspecified)

## Error: trying to use CRAN without setting a mirror

library(corrgram)

## Loading required package: seriation

corrgram(wineReordered, order = TRUE, lower.panel = panel.shade, upper.panel = panel.pie, 
    text.panel = panel.txt, main = "Predict price of wine")

## Error: object 'wineReordered' not found

# Linear Regression (one variable)
model1 = lm(Price ~ AGST, data = wine)

## Error: object 'wine' not found

summary(model1)

## Error: object 'model1' not found

# Sum of Squared Errors
model1$residuals

## Error: object 'model1' not found

SSE = sum(model1$residuals^2)

## Error: object 'model1' not found

SSE

## Error: object 'SSE' not found

# Linear Regression (two variables)
model2 = lm(Price ~ AGST + HarvestRain, data = wine)

## Error: object 'wine' not found

summary(model2)

## Error: object 'model2' not found

# Sum of Squared Errors
SSE = sum(model2$residuals^2)

## Error: object 'model2' not found

SSE

## Error: object 'SSE' not found

# Linear Regression (all variables)
model3 = lm(Price ~ AGST + HarvestRain + WinterRain + Age + FrancePop, data = wine)

## Error: object 'wine' not found

summary(model3)

## Error: object 'model3' not found

# Sum of Squared Errors
SSE = sum(model3$residuals^2)

## Error: object 'model3' not found

SSE

## Error: object 'SSE' not found

###### 

# Remove FrancePop
model4 = lm(Price ~ AGST + HarvestRain + WinterRain + Age, data = wine)

## Error: object 'wine' not found

summary(model4)

## Error: object 'model4' not found

# Correlations
cor(wine$WinterRain, wine$Price)

## Error: object 'wine' not found

cor(wine$Age, wine$FrancePop)

## Error: object 'wine' not found

cor(wine)

## Error: object 'wine' not found

# Remove Age and FrancePop
model5 = lm(Price ~ AGST + HarvestRain + WinterRain, data = wine)

## Error: object 'wine' not found

summary(model5)

## Error: object 'model5' not found

# Apply model on test set

# Read in test set
wineTest = read.csv("wine_test.csv")

## Warning: cannot open file 'wine_test.csv': No such file or directory

## Error: cannot open the connection

str(wineTest)

## Error: object 'wineTest' not found

# Make test set predictions
predictTest = predict(model4, newdata = wineTest)

## Error: object 'model4' not found

predictTest

## Error: object 'predictTest' not found

# Compute R-squared
SSE = sum((wineTest$Price - predictTest)^2)

## Error: object 'wineTest' not found

SST = sum((wineTest$Price - mean(wine$Price))^2)

## Error: object 'wineTest' not found

1 - SSE/SST

## Error: object 'SSE' not found