jdebacker / compecon_fall21 Goto Github PK

Create a function that returns the score. This function should take as arguments:

1. Arrays or data frames of buyer/target characteristics (corresponding to the matches on each side of the inequality)
2. Parameters (e.g., beta)
3. Maybe other optional keyword arguments

It should return a scalar that is the value of the score function.

There should be no loops inside this function.

Add descriptive docstrings to all functions. Please follow the Google docstring format. Keep all lines in docstrings (or in-line comments) to no more than 88 characters.

In PS5_solutions.py, write some nested for loops to create combinations of buyers and targets. The arrays will have 3 columns: year, buyerid, targetid. There were be four arrays: f(b,t), f(b',t'), f(b,t'), f(b',t). The rows should line up such that using the same row from all 4 arrays will allow you to evaluate: f(b,t) + f(b',t') > f(b,t') + f(b',t)

Note: looping over numpy arrays will be faster than looping over data frames.

Dear UofSC Research Computing Users,

UofSC Research Computing will present “Introduction to Linux Command Line”, the second in its series of Fall 2021 workshops, at 10:00AM EDT on Thursday, September 2, 2021, using Microsoft Teams. This workshop is open to all who would like to attend.

This workshop will introduce users who are new to Linux to a few basic commands which are required for successful use of the Hyperion supercomputer and Linux systems in general.

Please register for this workshop. Please mark the date and time of the workshop on your calendar. An Outlook invitation, including a link to the Teams meeting, will be sent to you in advance the workshop.

The Fall 2021 Research Computing Workshop Series is being presented in partnership with the University Libraries SHARPGrads program. Students are encouraged to also register for SHARPGrads.

Best Regards,

UofSC Research Computing
[email protected]

Create a function in PS5_solutions.py to create variables needed to determine the payoff matrices. This function should take a dataframe as an argument and then return a dataframe with all variable needed for analysis. Something like what is done on lines 10-15 of the current file, but perhaps with more variables created.

Dear UofSC Research Computing Users,

UofSC Research Computing will present “Introduction to Git”, the third in its series of Fall 2021 workshops, at 10:00AM EDT on Thursday, September 9, 2021, using Microsoft Teams. This workshop is open to all who would like to attend.

Git is a version control system that developers use to track and share code. This workshop will take you from a beginner to being highly proficient using Git and GitHub. You will learn how to collaborate on projects effectively with GitHub, build a solid foundation in Git, and master advanced version control skills.

Please register for this workshop. Please mark the date and time of the workshop on your calendar. An Outlook invitation, including a link to the Teams meeting, will be sent to you in advance the workshop.

The Fall 2021 Research Computing Workshop Series is being presented in partnership with the University Libraries SHARPGrads program. Students are encouraged to also register for SHARPGrads.

Best Regards,

UofSC Research Computing
[email protected]

To start on the extra credit assignment, someone needs to commit a file named PS5_solutions.py to PS5_solutions branch. This file should be in the ProblemSetSolutions directory.

cc @amshoulders, @Arpathamit , @mjakaria, @YingruHan , @Zilin-Xiong, @Joy-liningqiao

Creating an issue. Can someone respond that they see it?

PS5_Solutions.py now has a function called create_array_ids, which creates arrays of buyer and target ids that will be useful for creating the payoff matrices however, this function could be written in a better way:

1. There are multiple nested for loops over the same ranges. This can all be done in just one nested for loop
2. Use Numpy array objects -- not lists and DataFrames.

Update PS5_solutions.py to use a relative path to radio_merger_data.csv, which is in the Matching directory.

To ensure this works across operating systems, use os.path.join from the os library.