EC988 Data Analytics I: Mock Exam 2024 PDF
Document Details
Uploaded by Deleted User
University of Strathclyde
2024
University of Strathclyde
Tags
Related
- ECO1502 Introduction à la Macroéconomie PDF
- Data Revisions Are Not Well Behaved (2008) PDF
- Microeconomics for Managers (MEM) Lecture Notes PDF
- Wawancara Annual Report Award 2023 PT BANK NEGARA INDONESIA (PERSERO) TBK PDF
- Data Analysis for Business, Economics, and Policy PDF
- Eco1102S_3 PDF - Open Economy Macroeconomics
Summary
This is a past paper for the EC988 Data Analytics I course at the University of Strathclyde. The exam covers topics such as autoregressive models (AR(1)), vector autoregressions (VARs), and sign restrictions. The paper presents various questions to test the student's understanding and ability to apply these models to real-world economic scenarios.
Full Transcript
Department of Economics EC988 Data Analytics I: Essentials in Economics & Finance Attempt ALL Single Choice Question in Section A AND TWO of the THREE Questions in Section B Calculators must not...
Department of Economics EC988 Data Analytics I: Essentials in Economics & Finance Attempt ALL Single Choice Question in Section A AND TWO of the THREE Questions in Section B Calculators must not be used to store text and/or formulae nor be capable of communication. Invigilators may require calculators to be reset. PLEASE TURN OVER EC988 Page 1 of 6 Section A (30%) Answer ALL Single Choice Question in this section. Explain your choice with a brief statement! 1. Suppose 𝑦! evolves according to an AR(1) process: 𝑦! = a + r 𝑦!"# + e!. Here, a refers to the intercept, r to the autoregressive coefficient, and e! to the error term. Which parameter(s) determine whether the process is stationary, non-stationary, or explosive? [6%] a. a, b. r, c. The variance of e! , d. a and r. 2. Suppose 𝑦! evolves according to an AR(1) process: [6%] 𝑦! = a + r 𝑦!"# + e!. Here, a refers to the intercept, r to the autoregressive coefficient, and e! to the error term. If the intercept a = 0, what is the unconditional mean of the process? a. 0, b. r, # c. #"r , d. ¥. PLEASE TURN OVER EC988 Page 2 of 6 3. Suppose 𝑦! evolves according to a stationary AR(1) process. We aim to produce forecasts for this process. With respect to variance of these forecasts, which of the following statements is FALSE? [6%] a. The variance of the forecasts diminishes over the forecast horizon and approaches zero in the limit (i.e., as the horizon ℎ → ¥). b. In the short term, e.g., with monthly data, the forecast error variance of the two-month- ahead forecast is never smaller than the variance of the one-month-ahead forecast. c. The unconditional (steady-state) variance is constant. 4. Which of the following statements about the identification scheme of sign restrictions is TRUE? [6%] a. Sign restrictions identification is a rather “mechanical” procedure and does not necessarily yield economically meaningful impulse responses. b. The ordering of variables is important for sign restrictions. c. Sign restrictions use prior beliefs (typically in the form of economic theory) to define/restrict the signs of responses of endogenous variables to structural shocks. d. A characteristic of the sign restrictions identification scheme is that all impulse responses are typically zero on impact. 5. A typical variance-covariance matrix in a reduced-form vector autoregression (VAR): [6%] a. is symmetric and features strictly positive elements on the main diagonal. b. is a diagonal matrix. c. is a full (non-symmetric) matrix with non-zero off-diagonal elements. d. is symmetric and features non-zero (negative and positive) elements on the main diagonal. PLEASE TURN OVER EC988 Page 3 of 6 Section B (70%) ANSWER TWO of the THREE Questions in this section Q.2 Suppose we have quarterly data on the Federal Funds rate (𝐹𝐹𝑅), year-over-year inflation of consumer prices (𝐼𝑁𝐹), and the unemployment rate (𝑈𝑅) from 1960:Q1 to 2022:Q4. In the following, we aim to estimate the systematic component of monetary policy in the US. Specifically, we estimate an autoregressive distributed lag (ARDL) model in which the Federal Funds rate is the endogenous variable and both the inflation rate and the unemployment rate are treated as exogenous covariates. (a) Provide the ARDL regression equation, considering a single lag of the dependent variable (𝐹𝐹𝑅!"# ), the current values of the two exogenous covariates as well as the respective first lags (𝐼𝑁𝐹! , 𝐼𝑁𝐹!"# , 𝑈𝑅! , 𝑈𝑅!"# ). Carefully describe each of the parameters. [10%] (b) The OLS regression output of this model is provided below. Briefly discuss the relationship between Federal Funds rate and the explanatory variables. Comment on the significance of the parameters using a significance level of 5%. [10%] ======================================================================= (Intercept) 𝐹𝐹𝑅!"# 𝐼𝑁𝐹! 𝐼𝑁𝐹!"# 𝑈𝑅! 𝑈𝑅!"# ----------------------------------------------------------------------------------------------------------------------------- Estimate 0.3 0.9*** 0.2** -0.1 -1.4*** 1.3*** Std. Error (0.25) (0.02) (0.08) (0.08) (0.19) (0.19) ======================================================================= Note: *p