Submit the assignment by the due date via canvas. Assignments may be submitted up to 72 hours late for a 5 point reduction.
All files read, sourced, or referred to within scripts should be assumed to be in the same working directory (./).
Your code should be clearly written and it should be possible to assess it by reading it. Use appropriate variable names and comments. Your style will be graded using the style rubric [10 points].
Some of these exercises may require you to use commands or techniques that were not covered in class or in the course notes. You can use the web as needed to identify appropriate approaches. Part of the purpose of these exercises is for you to learn to be resourceful and self sufficient. Questions are welcome at all times, but please make an attempt to locate relevant information yourself first.
In this question you will demonstrate your understanding of important Linux shell skills by writing short shell commands to work with the 2015 RECS data. When instructed to use a “one-liner”, utilize pipes “|” and file redirection to format your answer as a single string of commands. If needed, assume the commands are being written in a Bash shell with only the default functionality on the servers in the login.itd.umich.edu pool.
Submit your answer to this question as a single shell script named ps1_q1.sh; also submit the same file with a txt extension ps1_q1.txt for viewing on Canvas. Use comments to clearly delineate each part. Also, be sure to include a descriptive header and “shebang” (#!/…).
[5 pts] Create a variable ‘file’ with the name of the csv file containing the RECS data. Check if this file exists in the local directory and, if not, download it.
[5 pts] Write a one-liner to extract the header row of the RECS data, translate the commas to new line characters, and write the results to a file ‘recs_names.txt’. [Hint: your solution should be a one-liner but you may wish to include in your script a code block that deletes recs_names.txt if the file already exists.]
[10 pts] Write a one-liner that uses ‘recs_names.txt’ to find the column positions for the id and replicate weight columns in the RECS data and then re-formats these positions as a single, comma-separated string. [Hint: For the final step, use the -s and -d options to the command paste.]
[5 pts] Store the result from the previous one-liner in the variable ‘cols’. Use this variable to write a one-liner that extracts the id and replicate weight columns from the recs data and writes them to recs_weights.csv. [Hint: In the first step, use a construction such as cols=$(...).]
In this question you will extend your knowledge of the Linux shell by modifying your solution to question 1 to write a short command line program to be named cutnames. This program should extract those columns from a csv file having headers matching a regular expression.
Your command/script should accept the arguments “file” and “expression” by position. It should reproduce the output from question 1 if called as below:
bash ./cutnames.sh ./recs2015_public_v4.csv 'DOEID|^BRR' > recs_weights.csvIt is also acceptable if it works instead if called as below:
bash ./cutnames.sh ./recs2015_public_v4.csv 'DOEID\|^BRR' > recs_weights.csvIn either case, be sure to include comments explaining how the command line arguments are used by the script.
Name your script cutnames.sh and, as before, submit to Canvas as both cutnames.sh and cutnames.txt.
Challenge: Modify your script to recognize and work with gzip compressed files when the file name passed has extension ‘.gz’. This is ungraded and should not be submitted.
In this question you will write several R functions for working with “mouse-tracking” data as described in this manuscript. Briefly, suppose you have data in the form of a series of triples (x, y, t) representing the position (x, y) of a mouse cursor in the plane (i.e. monitor) at time t during a trial in which participants click one of two buttons in response to a prompt. The buttons are arranged horizontally on opposite sides of the screen to allow the experimenter to garner information about how decisively each response is given.
For each part, write an R function to accomplish the stated task. Name each function using an informative verb. Use comments to document the arguments and output of each function. Be sure to clearly state any assumptions on the inputs.
Submit your answers to parts a-e as a single executable R script ps1_q3.R. Also submit a pdf created using Rmarkdown with answers to all parts including your function definitions.
[5pts] Write a function that accepts a n \(n \times 3\) matrix representing the trajectory (x, y, t) and translates it to begin with time zero at the origin.
[5pts] Write a function that computes the angle \(\theta\) formed by the secant line connecting the origin and the final position in the trajectory. Your answer should be an angle between \([-\pi, \pi]\). Be sure your your solution works for a trajectory ending in any of the four quadrants.
[5pts] Write a function to rotate the (x, y) coordinates of a trajectory so that the final point lies along the positive x-axis.
[2pts] Combine the three parts above into a single function that normalizes an \(n \times 3\) trajectory matrix to begin at the origin and end on the positive x-axis.
[5pts] Apply your function to the sample trajectories at the Stats506_F19 repo on GitHub and check your solutions against the sample measures. Then, compute the metrics above for the test trajectories and report your results in a nicely formatted table.