The Timlin Trust, Department of Economics, and the College of Graduate Studies are happy to announce the establishment of the Timlin Award for the best MA Project/Thesis in Economics in the preceding two years. The Award recognizes the most outstanding research MA Project/Thesis based on the contribution to research, originality of the idea, soundness of methodology, and quality of writing, as determined by the Award Committee. The Award carries a monetary value of $500.
Our 2020 Winner
Mila Markevych (MA, 2018)
Research Project: The Gender Wage Gap During the Resource Boom and Bust in Canada: What Effect Does Industrial Composition Have?
Supervisor: Prof. K. Foley
"This paper examines the effect of industrial composition on the gender wage gap during the resource boom and bust periods in Canada. Using Oaxaca-Blinder decompositions, I examine the effect of the distribution of men and women across industries on the wage gap in the resource extraction intensive region -- RE region (i.e. Saskatchewan, Alberta, and Newfoundland), and provinces that have a significantly lower concentration in the resource extraction industry -- Ontario and Quebec. The data for the analysis was taken from the Labour Force Survey for 1997-2017.
I find that these industrial composition effects account for part of the gender wage gap in all periods in both regions, with the magnitude of the effects being larger in the RE region. During the bust period, the magnitude of the industrial composition effects decreases in ON/QB, whereas it increases in the RE region. The slower convergence of male and female wages during the resource boom in the RE region can be explained, in part, by the resource extraction industry. During the bust period, the gender wage differential attributed to construction, agriculture, fishing and forestry industries increases in both regions, with the composition effects in these industries being higher in the RE region compared to ON/QB." - Mila Markevych
Our 2018 Winner
Mingshi Kang (MA, 2016)
Research Project: Size and Asymptotic Power Performances of HAR Wald Tests Using Fixed-b Theory
"The paper compares the performance of finite sample heteroskedasticity and autocorrelation robust (HAR) Wald tests using the traditional chi-squared critical values and fixed-b critical values introduced by Kiefer and Vogelsang (2005) in two and three restrictions tests. Simulation results show that using fixed-b critical values could substantially reduce size distortions in tests with ARMA errors. Prewhitening is suggested to use in tests using the Bartlett kernel while should not be used in tests using the QS kernel. An asymptotic power analysis is done for different combinations of kernels and bandwidths from a new perspective. Similar to the t-test results in Kiefer and Vogelsang (2005), when fixed-b theory applies, tests using the Bartlett kernel have higher powers and higher size distortions compared to those using the QS kernel. Both power and size distortion decrease as bandwidth increases. These results indicate the similar power and size trade-offs mentioned by Kiefer and Vogelsang (2005) in t-test.” –Mingshi Kang
Our 2016 Winner
Tofik Fite (MA, 2015)
Research Project: The Role of Quality of Skills in Explaining Immigrants’ Wage Gap in Canada
"The main objective of this project is to investigate the impact of differences in quality of skills in explaining wage differentials among immigrants and between immigrants and Canadians. The Blinder-Oaxaca decomposition technique is applied to the Ethnic Diversity Survey (2002), confidential microdata from Statistics Canada, accessed through the Saskatchewan Research Data Centre. The results reveal that the Sheepskin Effect - the premium associated with obtaining credentials – is higher for immigrants from traditional source countries such as US, UK, Australia, New Zealand, and North-Western Europe compared to immigrants from other regions. The endowment advantage, in terms of schooling and credentials, which immigrants in general have, is offset by unobserved productivity." - Tofik Fite
Our 2014 Winner
Lyndon Jacak (MA, 2013)
Research Project: Solving DSGE Models: A Comparison of the Neural Network and Log - BLinear Solution Models.
"The goal of this project was to explore the concepts and tools of analysis of dynamic stochastic general equilibrium (DSGE) modelling - specifically New Keynesian DSGE modelling - within a computer programming framework. The former was introduced to us rigorously in our MA coursework, but the opportunity to use software such as MATLAB to explore economic shocks and market failures was limited. For me, this project allowed an economic model to come alive." - Lyndon Jacak