BMRMSE-SM-QUAN: QUANTITATIVE RESEARCH METHODS
FEBRUARY 2022
RESIT FOR ASSIGNMENT 1
Welcome to the resit for the first individual assignment. This assignment consists of a series of 10 questions, testing knowledge from Sessions 1 and 2 of the course. Each question is worth three points, for a total of 30 points. This assignment is worth 30% of your final grade. You need 16.5 points to pass the assignment (equivalent to obtaining a 5.5). The highest result between the first assignment and this resit will count as the final grade for the first assignment.
In this assignment, you will:
1) Load up, summarize, and clean data into statistical software.
2) Run and interpret a linear regression model.
3) Think through important considerations for the empirical design of a quantitative study.
You will only require the commands and knowledge from class and the videos associated with these classes. All information required to complete the assignment is given within the assignment. Often, you will find some intermediary steps are required before certain commands work (for instance, due to issues with the underlying data). This is often the case in working with real-world data, and being able to fix this is an important step of the learning process. Make sure to keep track of your steps (for instance, via a .do and a log file) and to read the full information provided to you carefully.
It is recommended that you use Stata for this assignment, so make sure to have it downloaded and installed before you start. This assignment uses two datasets that are generated for you specifically via Stata; make sure to follow the set-up instructions exactly. Loss of points as a result of incorrect data generation will be your own responsibility!
You have one attempt to complete this assignment on Canvas, so make sure to have all your answers ready to fill in beforehand. Please DO NOT navigate away from the quiz window or close this window once you start it. Do not use any browser navigation buttons (i.e. Back, forward, Home etc). Please also be sure to submit your answers after completing the test.
I will provide detailed motivations for each answer after the deadline for this assignment has passed. There are two datasets that will be generated and used. Our datasets contain information on the following variables:
Dataset 1:
- firmid: An identifier with a unique value for each firm in the data. Numeric.
- year: The year which the observations is from. Numeric.
- industry: The industry of the firm (2-digit SIC code). Categorical, can take on any value between 1 and 5.
Dataset 2:
- firmid: An identifier with a unique value for each firm in the data. Numeric.
- firmyear: The combination of firmid and year with a “-“ in between. String.
- strategy: An indicator of whether or not the firm followed a specific strategy in the year. Dummy variable: either yes (1) or no (0).
- employees: The number of employees in the firm in the year (in FTE). Numeric, can take on any non-negative value.
- profits: The profits (or losses, if negative) of the firm in the year, in thousands of Euro. Numeric, can take on any value.
You can assume that the datasets are clean, in that the variables profits, employees, industry, and strategy do not require any editing, transformation, or cleaning.
1: Load up and combine the two datasets, merging dataset 1 to dataset 2—maximizing the extent to which the full data from the two datasets gets combined. Dataset 1 should be your “master” file, and dataset 2 your “using” file.
What kind of merge did you need to do?
2: We should always first assess whether our merge was successful: how many observations in the datasets were not matched? Is this what you would expect, given the nature of your match?
3: We are concerned about non-response bias for the strategy indicator. Specifically, we are worried that smaller firms will more often not report the relevant information to determine whether the firm engaged in the strategy or not. Do we observe any evidence of such non-response bias in our data? Why?
4: Next, you estimate two models. The first is your ‘baseline’ model, which predicts profits as a function of the number of employees and the industry of the firm. The second model is your main model of interest, where you estimate profits as a function of the number of employees, the industry of the firm, and the strategy indicator. Make sure to estimate these models with robust standard errors.
Note that for this assignment you do not need to estimate a panel data model. A simple linear regression will suffice, given that this assignment regard only the first two sessions. Answers using a panel data model will be counted as incorrect.
Create a regression table for these two models. Ensure proper “thesis level” formatting and quality. You can either paste the resulting, formatted, table directly into Canvas or can take a screenshot of the final table.
5: Interpret the coefficient for the employees variable in each of the models. This interpretation should build on the coefficients and their p-value to discuss the statistical significance and the meaning of the effects.
6: How can the effect(s) of industry be interpreted, intuitively (i.e., no need to report numbers)? Here, focus on the main model of interest (that is, the one also including the strategy indicator).
7: Suppose that we want to test the hypothesis that the strategy which we are interested in has a positive effect on firm performance. Do we find support for our hypothesis? Why (not)?
8: Do you think we can make a causal claim about the above hypothesis, given our relationship of interest and our empirical design? Why (not)? What specific challenges are (or are not) present?
9: We also want to know whether the explanatory variables—jointly—have a statistically significant effect on the outcome variable in our main model of interest. Which statistic(s) can we use to investigate this? What is the value of this statistic, and what are its degrees of freedom? Where do these degrees of freedom come from?
10: Calculate the predicted value for a firm in industry number 1, with five employees, and with the strategy implemented using the main model of interest. Compare this value to that of a firm in industry number 1, with five employees, and without the strategy implemented. How large is this difference and what does the difference between these two values mean? How does it relate to the coefficients from the model?
CS 340 Milestone One Guidelines and Rubric Overview: For this assignment, you will implement the fundamental operations of create, read, update,
Retail Transaction Programming Project Project Requirements: Develop a program to emulate a purchase transaction at a retail store. This
7COM1028 Secure Systems Programming Referral Coursework: Secure
Create a GUI program that:Accepts the following from a user:Item NameItem QuantityItem PriceAllows the user to create a file to store the sales receip
CS 340 Final Project Guidelines and Rubric Overview The final project will encompass developing a web service using a software stack and impleme
