Material

🎥 Watch Models with Multiple Predictors

• Slides

Today's Goal

• Use functions in R to fit a linear model with multiple predictors
• Interpret coefficients of the model
• Understand and compare $R^2$ and adjusted $R^2$
• Model interactions between variables

Quiz

Q - Which of the following is the correct code to fit a linear model for $y$ with $x_1$ and $x_2$ from mydata?

a. linear_reg(engine = "lm") %>% fit(y ~ x1 & x2, data = mydata)

b. linear_reg(engine = "lm") %>% fit(y ~ x1 ~ x2, data = mydata)

c. linear_reg(engine = "lm") %>% fit(y ~ x1 + x2, data = mydata)

d. linear_reg(engine = "lm") %>% fit(y ~ x1 * x2, data = mydata)

Quiz

Q - Which of the following is the correct code to fit a linear model for $y$ with $x1$ and $x2$ from mydata?

a. linear_reg(engine = "lm") %>% fit(y ~ x1 & x2, data = mydata)

b. linear_reg(engine = "lm") %>% fit(y ~ x1 ~ x2, data = mydata)

c. linear_reg(engine = "lm") %>% fit(y ~ x1 + x2, data = mydata)

d. linear_reg(engine = "lm") %>% fit(y ~ x1 * x2, data = mydata)

Quiz

Q - What is the biggest difference in interpreting regression coefficients of a model with multiple predictors compared to a model with a single predictor?

a. "all else at 0"

b. "all else held constant"

c. "all increased by 1 unit"

d. no difference

Quiz

Q - What is the biggest difference in interpreting regression coefficients of a model with multiple predictors compared to a model with a single predictor?

a. "all else at 0"

b. "all else held constant"

c. "all increased by 1 unit"

d. no difference

Quiz

Q - What does Occam's razor state in terms of model selection?

a. "choose the simplest model"

b. "choose the model with more predictors"

c. "choose the model with the highest predictive power"

d. "choose the parsimonious model"

Quiz

Q - What does Occam's razor state in terms of model selection?

a. "choose the simplest model"

b. "choose the model with more predictors"

c. "choose the model with the highest predictive power"

d. "choose the parsimonious model" the simplest best model

Quiz

Q - Which of the following is the value to consider when comparing multiple models? Select all that apply.

a. Type I error

b. Adjusted R-squared

c. Coefficients

d. R-squared

Quiz

Q - Which of the following is the value to consider when comparing multiple models? Select all that apply.

a. Type I error

b. Adjusted R-squared

c. Coefficients

d. R-squared

Let's Practice Together!

Go to AE 21: Models with Multiple Predictors 1

Bulletin

• Watch videos for Prepare: June 13

• Lab07 due tonight at 11:59pm

• Project draft due Monday, June 13 at 11:59pm

• HW02 due Thursday, June 16 at 11:59pm.

• HW04 due Thursday, June 16 at 11:59pm

• Submit ae20 and ae21