Title | Econ2206 notes 1.1 |
---|---|
Author | Chaitanya Narayan |
Course | Introductory Econometrics |
Institution | University of New South Wales |
Pages | 2 |
File Size | 81.8 KB |
File Type | |
Total Downloads | 59 |
Total Views | 157 |
Econ2206 notes for lecture 1.1...
Econ2206: Introductory Econometrics Week 1: Administration and Introduction Assessment dates o 2 Problem sets Due: 8th October, November 27th o Midterm 28th October o Online Quizzes Week 2, 4, 7, 10 Check out STATA from MyAccess
Econometrics The use of statistical methods to analyse economic data Typical goals o o o o
Estimating relationships between economic variables Testing economic theories and hypotheses Forecasting economic variables Evaluating and implementing government and business policy
Example: We have data for different countries and years of the unemployment rate and inflation The goals here are:
Estimating the relationship: When employment is high is inflation also high? Testing an economic theory: High inflation causes employment to go up Forecasting: How high inflation will be in the next years Evaluation: If the AUS Govt increases taxes what happened to unemployment
Economic Data Types of data:
Cross-sectional o Sample of individuals, households, firms at a given point of time/period o Observations typically more or less independent E.g. pure random sampling Census data is often random sampled and cross sectional Time series o Observations for an individual, firm, country,… collected OVER TIME E.g Stock prices, exchange rates, consumer price index, GDP
They are typically serially correlated Observations between themselves are not always dependent but there is some relationship o Trends, cycles, seasonality o Data Frequency (how often we observe the variable) Pooled cross section o 2 or more cross sections combined in one data set o Used to evaluate policy changes Effect of change in property tax on house prices Random sample in 1993 and Random sample in 1995 The 2 samples contain different observation sets (essentially pre VS post) Panel/Longitudinal data o Same cross-sectional units followed over time o Combines cross-sectional and time series dimension o Very useful in modelling E.g. City crime statistics: each city is observed in 2 years E.g. effect of police on crime rates may exhibit time lag
Random Variable VS Sample Random variable: Function that describes the possible outcomes and probabilities of a random event. E.G. A coin toss. We always write random variables in capital letters – X, W, or COIN Random sample: The extrapolation of that random variable. I.E actually flipping the coin and recording observations of the first 10 tosses. {w1 , w2 ,… , w 10 } We always write random samples in lower case – w, x, coin...