BSB123 Data Analysis/ECOM30003 Applied Microeconometric Modelling/STAT2402 Analysis of Observations/DATA2002 DATA2902/EGB120 Foundations of Electrical Engineering

2nd September

Question One (9 Marks)
Aptitude / psychometric tests are common tools used by the public service and many private employers to gauge employees suitability. One such test used by a large company had an average score of 140 with a standard deviation of 16. If the scores on the test were normally distributed find:
Probability a candidate scored less than 100.
Probability of getting more than 120
Probability of getting between 150 and 180
If only the top 5% of candidates are invited for interview, what score would you need to achieve to get invited for one? (Check: 166.318)
What is the interquartile range for the test?
What is the probability that two strangers sitting side by side would both score more than 170? (Check: 0.000924)
Draw diagrams for each of parts a. to d. above

Question Two (8 Marks)
8% of people who buy theatre tickets do not turn up. In a sample of 150 theatre goers, find the probability that:
Exactly 12 do not turn up to the next show they have tickets for.
Less than 10 don’t go.
15 or more don’t turn up.
Between 7 and 11 inclusive do not turn up.
More than 140 DO turn up (Check: 0.2315)
Calculate the expected value and the standard deviation of the number who do not show up. (Check: σ = 3.323)

Question Three (4 Marks)
Many events (concerts, festivals etc) are ticketed, but do not have specific seating. For such events there is usually a maximum venue capacity, however, it is possible to oversell the event because on many occasions people do not turn up despite purchasing tickets.

One such event, A Day on the Grass, has a notional capacity of 750 patrons, however for past events just on 12% of ticket holders do not turn out.
What is the probability the event does not exceed maximum capacity if the venue sold 850 tickets? (Check: 0.599)
How many tickets could they need to sell in order to ensure less than a 1% chance they did not exceed capacity? (Note this question requires some trial and error)

===================================================================================================================================

ECOM30003: Applied Microeconometric Modelling

Assignment 1

This assignment is based on ''Can Electronic Procurement Improve Infrastructure Provision? Evidence from
Public Works in India and Indonesia" by Lewis-Faupel, Neggers, Olken and Pande. The paper conducts an
evaluation of the impact of e-procurement in India and Indonesia on outcomes related to the process, as
well as outcomes related to cost and quality of projects undertaken under government procurement.
Your submission should include your written responses to all questions, and then an appendix including all
requested tables, and your stata code. Tables should be numbered sequentially. The table containing results
for Questions 7-11 should have column headings indicating the question they correspond to. Please keep
your written answers brief and to the point, referring to the tables by table number and column heading
where appropriate.
1. [2 marks] What is the empirical challenge in identifying the causal impact of the e-procurement
program.

2. [7 marks] Briefly explain the strategy that the authors employ to identify the causal impact of e-
procurement and the empirical specification used to implement this strategy. What is the parameter

of interest? What is the interpretation of this parameter?
3. [2 marks] What is the key assumption that must be met for this approach to deliver estimates of
causal effects of e-procurement?
4. [3 marks] Is the error term in the model above likely to be identically and independently distributed
(iid)? If not, why not? Is this a problem for inference? How should the standard errors be estimated?
Justify your answer.
Questions 5 and 6 asks you to look at variation in the data that will be used to identify the policy impact.
5. [6 marks] This question asks you to produce a table using state level data showing the year in which
each state introduced e-procurement. The objective is to show the year and states that we
potentially have package level data to estimate treatment effects.
Step1: Generate a new variable that is zero if e-procurement is introduced after the years for which
there is information on procurement packages, and equal to the year that e-procurement is
introduced otherwise. [Hint: tab package_year to see the years for which there is information on
procurement packages].
Step 2: Use the collapse command to create a state level data set containing the variable you just
created. {Hint: use stata’s help to find out how to use the collapse command.]
Step 3. Using the collapsed data set, create a 2 way tabulation, also called a crosstabulation,
showing the year e-procurement was introduced in columns and states in the rows of the table. The
first row should look like the example below. Label the table Table 1.
state_id 0 2004 2006 2007 2008 Total

Using Table 1 answer the following questions. How many states are there in the data? How many
states introduced e-procurement during the years for which we have data on procurement?
Describe the timing and number of states introducing e-procurement.
6. [6 marks] This question digs a bit more deeply into the data available to estimate treatment effects.
Specifically, we want to show the years and states for which e-procurement was actually used in the
sample of procurement packages that the authors use to estimate their models.
Step 1. Create the sample used in Table 4 column 6 of Lewis-Faupel et al. [Hint: An easy way to

work out observations used in this sample is by estimating the model and using stata’s post-
estimation command e(sample) to define a sample used in estimating the model contained on Table

4 column 6. Use stata’s help to find out about the e(sample) command.]
Step 2. Using this sample, create a table cross tabulating e-procurement status (ie whether the
policy is used for any packages in a state) and state. Show e-procurement status in columns and
states in the rows of the table and label the table Table 2. See below for the first row of Table 2.
state_id eproc=0 eproc=1 Total
Use Table 2 to answer the following questions. Comment on the percent of observations in the
sample used to create Table 2 for which the policy is in use. In Table 1 you identified all the states
that introduced e-procurement during those years we have procurement data. Are they all
represented in Table 2? If not, why is this the case?
Questions 7—11 asks you to estimate various models. Please report (all) results in a single table labelled
Table 3, and use the same format as Table 4 of Lewis-Faupel et al. Also, to rule out the potential for
different sample sizes driving different estimates, make sure you use the sample used in Table 4, panel A,
column 6 of Lewis-Faupel et al. The only specification for which the sample size should be different is the
specification estimated in Question 11. The standard errors reported for all specifications should account for
any issues identified in Question 4. [Hint: See outreg2 or esttab to output stata results into excel.]
7. [5 marks] Regress the outcome used in Table 4 panel A column 6 on the indicator for e-procurement
(without any other controls). What is the interpretation (sign, size and significance) of the estimated
coefficients?
8. [6 marks] Add indicators for year of package to the above specification. What happens to the
estimated coefficient on e-procurement (size and significance)? What does this tell you about the
change in the outcome variable over time? In light of this, do you think the specification from
Question 7 can be used to obtain the causal impact of e-procurement? Why?
9. [6 marks] Add state fixed effects to the specification estimated in Question 8. What happens to the
estimated coefficient on e-procurement (sign, size and significance)? In light of this, do you think
the specification from Question 8 above can be used to obtain the causal impact of e-procurement?
Why?
10. [5 marks] Estimate the specification reported in Table 4, panel A column 6. Explain why the year of
first inspection and monitor fixed effects are included. Compare the estimated coefficient on e-

procurement (sign, size and significance) from this specification to the specification estimated in
Question 9.
11. [6 marks] Re-estimate the specification estimated in Question 10, but exclude observations from the
state Andhra Pradesh. What happens to the estimated coefficient on e-procurement (sign, size and
significance) when you drop these observations from the estimation sample? What do you
conclude?


===================================================================================================================================

STAT2402 Analysis of Observations
11 September 2022

Information on the Data

The dataset is on 1063 randomly selected house in Saratoga suburb in New York. The data
contains the sale prices of the house and variables that describe the property, as detailed below.
Analyse the data to determine the relationship between selling price (Price) and the other
variables in the data.
The variables are as follows.
Price: The selling price of the house in $,000
Size: The floor area of the house in hundreds of square feet
Baths: The number of bathrooms
Bedrooms: The number of bedrooms
Fireplace: 1 = yes, 0 = no
Acres: land size
Age: Age of house in years


===================================================================================================================================

DATA2002 DATA2902

3.1 Example YAML
The following YAML code can be used to make sure you meet the minimum criteria. The self contained and
code folding options are particularly important.
3.1.1 R Markdown
If your file ends in .rmd you can adapt this:
title: "Your title here"
date: "`r Sys.Date()`"
author: "Your SID (don't put your name, so that we can respect the anonymous marking policy)"
output:
html_document:
self_contained: true # Creates a single HTML file as output
code_folding: hide # Code folding; allows you to show/hide code chunks
code_download: true # Includes a menu to download the code file
toc: true # (Optional) Creates a table of contents!
toc_float: true # table of contents at the side
number_sections: true # (Optional) Puts numbers next to heading/subheadings
---
3.1.2 Quarto
If your file ends in .qmd you can adapt this:
---
title: "Your title here"
date: "`r Sys.Date()`"
author: "Your SID (don't put your name, so that we can respect the anonymous marking policy)"
format:
html:
self-contained: true # Creates a single HTML file as output
code-fold: true # Code folding; allows you to show/hide code chunks
code-tools: true # Includes a menu to download the code file
table-of-contents: true # (Optional) Creates a table of contents!
number-sections: true # (Optional) Puts numbers next to heading/subheadings

===================================================================================================================================

EGB120 Foundations of Electrical Engineering

Rick the Raving Renewables Researcher realises his home Wi-Fi modem-router is on 24 hours a
day, every day. As it is powered from a 12 V 2 A power supply, Rick decides he should instead
power it from a solar panel on his roof. A (nominal) 12 V rechargeable battery will capture the
excess energy during the day, and keep the router running through the night. Then as well as
saving energy, Rick knows he can watch the live-streamed EGB120 lecture at his home on his
laptop even if the power fails!
Rick places the rechargeable 12 V 100 Ah battery downstairs outside the house. The Wi-Fi router
is in Rick’s upstairs study. Rick adds a (nominal) 12 V LED light above it so there will always be
light in the study. Rick mounts a second hand solar panel on the roof to get the best sunlight.
Rick connects these components together with spare “Cat 5” network cable. Each wire has a
resistance of about 0.1 Ω per metre, so Rick uses all eight wires in the cable, four in parallel for
each connection. The lengths of wire between the different items are shown in the diagram
below: the solar panel is c m from the LED light, which is 2 m from the Wi-Fi router, which is
then d m from the battery.
From Rick’s electrical engineering knowledge, he knows that the solar panel acts principally as
a current source with some parallel resistance and the Wi-Fi router as a constant current sink.
The LED light and the battery can each be modelled as a voltage source in series with some
internal resistance. The circuit of the system is shown below. Rick will measure the actual
values of these different components and promises to email these values to you separately.


(a) Rick decides to start simple and initially does not connect the solar panel. Redraw the circuit
with just the LED light, router, and battery. Correctly annotate the circuit components with the
values you have been emailed. These are the actual values Rick has measured, which you should
use in your calculations. Where possible, simplify and redraw the diagram by combining series
and parallel resistances.

(2 marks)
(b) Using a circuit analysis technique of your choosing, calculate the mesh currents and power
balance of the system running just from the battery. Show all working. Comment on the
effectiveness (calculate percentages) of the power transfer from the battery to the light and the
router. What power is dissipated in the wiring?

(4 marks)
(c) Use circuit simulation software to verify your calculations in (c). Ensure you only use linear
circuit elements (resistors, voltage and current sources) so your simulation exactly matches your
circuit and calculations. Capture screenshots of the software showing displayed values of
voltages and currents, and then confirm your results.

(2 marks)
(d) Rick charges the 100 Ah battery fully before connecting the LED light and Wi-Fi Router. Calculate
the stored charge in the battery in Ah after one day (i.e. 24 hours) has passed.

(2 marks)
(e) Rick now connects the solar panel to the circuit. Redraw the circuit to include the solar panel
and associated wiring, and simplify. Using a circuit analysis technique of your choosing, calculate
the mesh currents and power balance of the system at midday, when the solar panel is delivering
its rated current. Show all working. What current is flowing into the battery?

(4 marks)
(f) Rick is wondering if his solar panel is big enough to keep his system running continuously.
Calculate the minimum number of hours of equivalent full sun (known as “Peak Sun Hours”, PSH),
and thus rated solar output needed each day to replace the charge used during the remainder of
the day and overnight.

(2 marks)
(g) Rick is planning to upgrade his router to a new unit that will draw more power. However, the
router input voltage must remain above a minimum of 10.0 volts. Disconnect the router, and
calculate the night time Thévenin equivalent circuit at the point where the router connects.
Based on this Thévenin equivalent circuit, calculate the maximum current the new router can
draw. Note: for the night time equivalent circuit, you should set the solar panel current to zero, but
its shunt resistance should remain connected.

(3 marks)