Review

1. Generate a sequence of the numbers 10,9,8,7,6,5,4,3,2,1

2. Extract only numbers divisible by 3 from the sequence generated in Q1.

3. Generate the sequence 1,2,3,4,1,2,3,4,1,2,3,4

4. Generate the sequence 1,1,1,1,2,2,2,2,3,3,3,3

5. Replace all odd numbers in Q1 with their square and leave the even numbers the same

6. Generate all sliding windows of length 4 from the sequence in Q1. The first vector is 10.9,8,7 and the last one is 4,3,2,1

7. Generate the matrix shown below and find the row and column sums

1	2	3	4
5	1	7	8
9	10	NA	12
13	14	15	1

8. Scale the matrix from Q7 so each row has mean of 0 and standard deviaion of 1.

9. Generate and assign row names pt-1, pt-2, pt-3, pt-4 and column names gene.1, gene.2, gene.3, gene.4 to the matrix from Q7.

10. Convert the matrix from Q7 to a data.frame and add a column group with values A,B,A,B.

11. Remove the row with a missing value (NA) from the data.frame in Q10.

12. Find the average value of each gene by group using the data.frame from Q11.

13. Reshape the data.frame from Q10 to have only 3 columns group, gene, value. The geen column should have entries such as gene:1.

14. Sort the data.frame from Q11 in decreasing order of gene.1.

15. Replace all missing value with the column mean for the data.frame from Q10. group. Create a new data.frame that contains only the log values for all genes and the group column.

16. create a new data.frame with columns for genes.5, genes.6, rows for pt-2, pt-3, pt-1, pt-4 (in this order) and values drawn from a Poisson distribution with rate 10. Merge this with the data.frame from Q10 to get a new data.frame with 4 rows and 7 columns.

17. Generate data using the code shown below. Then fit a linear model to the data and print the coefficients and associated p-values for x1, x2, x3.

set.seed(123)
n <- 10
x1 <- runif(n, 0, 10)
x2 <- runif(n, 0, 10)
x3 <- runif(n, 0, 10)
y <- 2 + 0.5*x1 + 0.05*x3 + rnorm(n)
df <- data.frame(y=y, x1=x1, x2=x2, x3=x3)

18. Fit a linear model to the data below. Explain the results.

set.seed(123)
n <- 10
x1 <- 1:10
x2 <- seq(2,20,by=2)
x3 <- seq(3,30,by=3)
y <- 2 + 0.5*x1 + 0.05*x3 + rnorm(n)
df <- data.frame(y=y, x1=x1, x2=x2, x3=x3)

19. Load the data set at https://www.openintro.org/stat/data/bdims.csv into a data.frame using read.csv. Use knn with 5 neighbors on the wgt and hgt columns to predict the sex and generate a classification table of true and predicted values from LOOCV.

20. Using the same data set from Q19, perform a linear regression to predict the age from the 5 variables most correlated with the age. Use LOOCV correctly to get predicted age for each subject. Calculate the root mean square (RMS) error (square root of the mean of squared residuals) from the LOOCV predictions. Plot the predicted against the observed ages.