Economics Exam Paper May 2016
Uploaded by
Judhajeet ChoudhuriEconomics Exam Paper May 2016
Uploaded by
Judhajeet ChoudhuriCommon questions
Powered by AIThe coefficient of determination, R², equals the squared correlation coefficient between the dependent and independent variables only if there is an intercept in the regression equation because the presence of an intercept ensures the regression line passes through the mean of the data. This guarantees that the total variance is correctly partitioned into explained and unexplained components, aligning R² with the squared correlation. Without an intercept, the mean errors do not necessarily sum to zero, which distorts this relationship .
The inclusion of a disturbance term in a two-variable linear model is essential as it captures the effect of all omitted variables that influence the dependent variable, assuming they are uncorrelated with the included regressors. It accounts for random variability not explained by the model, ensuring that the OLS estimators remain unbiased and consistent under the classical assumptions. Omitting the disturbance term would lead to biased and inconsistent estimates if other relevant variables are correlated with the included ones .
Lewis' model posits that economic development occurs as surplus labor is transferred from the agricultural to the industrial sector where marginal productivity is higher, thereby boosting overall productivity and income levels. However, it faces criticisms such as underestimating the absorptive capacity of industry, disregarding rural sector innovation and productivity, and assuming infinite labor supply which doesn't consider demographic transitions. Critics also highlight the lack of attention to labor rights and equitable income distribution .
The Durbin-Watson statistic may be inappropriate for the demand for money analysis if the model includes a lagged dependent variable as one of the regressors, yielding an inherent bias since the DW test assumes residuals to be uncorrelated with each other. In models like this demand function which often measure economic behavior over periods, the DW statistic loses its power, making alternatives such as the Breusch-Godfrey test more suitable for detecting autocorrelation in dynamic settings .
The Ordinary Least Squares (OLS) estimator is considered the Best Linear Unbiased Estimator (BLUE) under the Gauss-Markov assumptions which require linearity in parameters, random sampling, no perfect multicollinearity, zero mean of disturbance, homoscedasticity, and no autocorrelation. The estimator remains unbiased and exhibits minimum variance among all linear unbiased estimators due to the fulfillment of these conditions. Deviations from these assumptions can affect the unbiased nature or efficiency of the OLS estimator, necessitating alternative estimation techniques such as Generalized Least Squares in the presence of heteroscedasticity or autocorrelation .
Asymmetric information is relevant as it challenges the efficacy of government interventions by potentially leading to government failures if policies are based on incomplete or skewed information. It can distort market outcomes by causing adverse selection or moral hazard, resulting in inefficient allocations such as inflationary financing or misallocated subsidies. Assessing information asymmetry is crucial for designing policies that improve market operations without unintentionally exacerbating market failures .
The White test detects heteroscedasticity by first estimating the regression model, then regressing the squared residuals on the original regressors, their squares, and their cross-products. A significant F-statistic in this auxiliary regression indicates heteroscedasticity. Detection is important as heteroscedasticity violates the OLS assumption of constant variance of errors, leading to inefficient estimates and invalid hypothesis tests. Addressing it ensures valid inference and policy implications drawn from the model .
In a simple linear regression model estimated using Ordinary Least Squares (OLS), the covariance between the estimated errors and the regressor is zero by construction when an intercept is included. This is because the OLS method minimizes the sum of squared residuals, inherently making the residuals uncorrelated with the regressors. Without an intercept, the covariance does not necessarily equal zero since the mean of the errors may not sum to zero, resulting in potentially correlated errors. This distinction is crucial for correctly interpreting regression outcomes and ensuring unbiased coefficient estimates .
The Harris-Todaro Model explains urban migration by considering the expected rather than actual wages, suggesting that would-be rural migrants decide based on potential urban employment, adjusted for unemployment probability. This results in excessive migration when urban unemployment still yields higher expected income than rural jobs. Economic implications include urban job competition, wage pressure, and possible urban poverty spikes if urban job creation lags behind migration. Policy measures must address wage equalization, employment opportunities, and rural development to stabilize migration flows .
Distinguishing between fixed rent and sharecropping contracts is crucial in modeling economic data because each contract type has different implications for efficiency and incentives. Fixed rent contracts require tenants to pay a set rent, promoting productivity if tenants can keep all additional revenue. Sharecropping involves tenants giving a portion of their output to the landlord, aligning interests but potentially reducing effort due to reduced marginal benefit from extra labor. These different structures impact the distribution of income and resources, affecting economic behavior and policy outcomes .