CHAPTER -1 1. Concepts of Econometrics 2. Types and Illustration of Econometric Data 3. Modeling and Forecasting 4. Methodology of Econometrics 5.

Citreous Paribus and undefined variables 6. Elementary Statistics 7. About Econometric Software ------------------------------------------------------------1. CONCEPTS OF ECONOMETRICS Meaning & Definition of Econometrics: Econometrics means measurement of economic concepts and indicators. The econometrics may be defined as the social science in which the tools of economic theory, mathematics and statistical inference are used in the analysis of economic phenomena. Econometrics is an amalgamation of economics, mathematical and statistics properties used in the structural, empirical and policy study of experimental and non-experimental data in scientific manner. Scope of Econometrics in Economy & Business : Econometrics is widely used in the empirical, structural and policy analysis of macroeconomic (open economy & national economy), microeconomic (market, firm and individual) and non-economic data. Here we cover study some of the macroeconomic variables (data of financial and commodity market) and non economic variables related to the individual, household, society, firm, market and country e.g. crime data, literacy data etc. Example of macroeconomic data of individual, market (financial and commodity), aggregate

economy and open economic data, comprises Individual Forecasting of consumption/investment pattern. Firm Forecasting of consumption/investment pattern of firm. Market Analysis of forecasting of operational behaviour of both market, financial market and commodity market, e.g. - Credit market, Capital market Money market and Commodity Market. Credit market- Analysis and forecasting of interest rate, credit/deposit and related behaviour. Capital market - Analysis and forecasting of quantity movement, price movement and riskreturn behavior of stock). Money market - Analysis and forecasting of monetary instruments, repo rate and reverse-repo rate operation. Commodity Market Analysis and forecasting of quantity movement, price movement & risk-return of goods. Aggregate Economy - Analysis of aggregate economic indicator, e.g. GDP, GNP, NNP, Employment, general price. Open Economic Macroeconomics Analysis and forecasting of open economic variables e.g. EXIM with US and G-8 etc. Non-Economic Variables Analysis and forecasting of all non-economic variables of country such as social, legal, political, ecological, environmental, geographical, global, technological. Types of Econometric Analysis : Empirical Analysis: In an empirical analysis we use data to test a theory or to estimate a relationship between results of data and predefined theory. Structural Analysis: In structural analysis we use data having definite structure to test a theory and estimate a structural relationship between variables. Policy Analysis: In policy analysis we use data to form a policy for government and business. 2. TYPES AND ILLUSTRATION OF ECONOMETRIC DATA : A. Types of economic data for modeling & forecasting

Single variable data Single independent variable. Two variable data two independent variables. Multiple variable data - two or more independent variables Serial data (single/two/multiple) large series of data with serial var., two var., or multiple var. B. Types of economic data on the basis of structural properties Cross-sectional data data specify point of time. Time Series data (single/two/multiple) data specify period. Pooled Data mix nature of cross-sectional and time series.

C. Experimental & Non-Experimental Data In econometric analysis a researcher uses Non-experimental Data as well as experimental data. (A non-experimental or experimental data may be of cross sectional type, time series type or pooled type). Non-experimental Data: The data which are not collected through controlled experiment on individual, firms or segment of economy. Non-experimental data is also known as observational or retrospective data. Experimental Data: The data which are collected through controlled experiment in laboratory environment in the natural sciences. Illustration of economic data on the basis of structural properties: Cross-Sectional Data : wages and other individual characteristics Observation 1 2 3 4 5 525 Wages 3.10 3.24 3.00 6.00 5.30 11.56 Educ 11 12 11 8 12 16 Exper 2 22 2 44 7 5 Femel 1 1 0 0 0 0 Married 0 1 0 1 1 1 3

526

3.50

14

TIME SERIES : Minimum wages, unemployment and related data of Costa-Rica Observation 1 2 3 525 526 Year 1950 1951 1952 1986 1987 avgmin 0.20 0.21 0.23 3.35 3.35 Avgcov 20.1 20.7 22.6 58.1 58.2 Unemp 15.4 16.0 14.8 18.9 16.8 gnp 878.7 925.0 1015.9 4281.6 4496.7

POOLED DATA : Two Year Housing Price Observation 1 2 3 250 251 252 253 520 Year 1993 1993 1993 1993 1995 1995 1995 1995 hprice 85500 67300 134000 243600 65000 182400 97500 57200 proptax 42 36 48 41 16 20 15 16 sqft 1600 1440 2000 2600 1250 2200 1540 1100 bdrms 3 3 4 4 2 4 3 2 bthrms 2.0 2.5 2.5 3.0 1.0 2.0 2.0 1.5

Note : Details in Chapter 1 of the handout. 3. MODELING AND FORECASTING Essentials Components of Model Derivation 1. Dependent Var., explained var., response var., predicted var., or regressand. 2. Independent var., explanatory var., control var., predictor var., regressor or covariate. 3. Constant Value or Intercept parameter; 4

4. Coefficient value or Slope parameter; 5. Error Term or Disturbance or Unobserved Variables or Residual; 6. Dummy Variable, Binary Variable or Qualitative Variable.

Explanation of econometric components Yi = 0 + 1Xi + i var., or regressand. 2. In equation (i) Xi is Independent var. also known as explanatory var., control var., predictor var., regressor or covariate. 3. In equation (i) 0 is Constant Value also known as Intercept parameter. 4. In equation (i) 1 and 2 is Coefficient value also known as Slope parameter; 5. In equation (i) i is Error Term also known as Disturbance or Unobserved Variables or Residual. 6. In equation (ii), the notation D is known as Dummy Variable also known as Binary Variable or Qualitative Variable. Yi = 0 + 1Xi + 2D + i Forecasting Ythat = t-1hat + t-1hatXt ------------- (iii) ------------- (ii) ------------- (i)

1. In equation (i) Yi is dependent variables also known as explained var., response var., predicted

For forecasting of Y value for the time t we need independent value of X of time t, coefficient value of time t-1 and constant value of time t-1. Yt hat is predicting the value of desired time t which is based on the independent data value Xt of the same year as well as coefficient and intercept value upto last time t-1. In next chapter, we will discuss forecasting of data in detail. 4. METHODOLOGY OF ECONOMETRICS 1. Statement of theory or hypothesis 2. Specification of mathematical model of the theory 3. Specification of econometric model of the theory

4. Obtaining the data 5. Estimation of the parameters of the econometric model 6. Hypothesis testing and testing of significance of parameters 7. Forecasting or prediction 8. Using the model for control or policy purpose. Explanation of Econometric Methodology 1. Statement of theory or hypothesis : Derivation of model in form of statement or hypothesis Economic Model of Crime (Model of Grey Baker-1968), describes Crime is the result of economic and socio-economic variables. 2. Specification of mathematical model of the theory Specification of mathematical model of the theory by using mathematical tools. Y = f (X1, X2, X3, X4, X5, X6, X7), (1.1.) Where : Y = hours spent in criminal activities X1 = wages for an hr. spent in criminal activities X2 = hourly wage in legal employment X3 = income other than from crime or employment X4 = probability of getting caught X5 = probability of being convicted if caught X6 = expected sentence if convicted, and X7 = age. Further, economic model of wages states wage is the result of education, experience and training. Or wages = f (edue, exper, training), Or X1 = f (w, e, t), where : (1.2a) .. (1.2)

wage or w = hourly wage edue or e = year of education training or t = week spent in job training. Note : function f(.) describes independent variables and parameters. 3. Specification of econometric model of the theory Specification of econometric model of the theory by using mathematical tools. Y = f { + (1X1, 2X2, 3X3, 4X4, 5X5, 6X6, 7X7,) + }, .. (1.3) Where : is intercept or constant 1 through 7 are the coefficient, X1 through X7 are independent variables, and is the stochastic error term, residual or unobserved variable. 4. Obtaining the data for dependent variable Here dependent variables is Y say Crime. 5. Estimation of the parameters of the econometric model here parameters are : a. Constant b. Coefficient (all 7 variables) 6. Estimation Error () and of Fitted value of Y (Yhat) 7. Test of significance of parameters (constant and coefficients) with help of variance error and standard error (Discussion in next chapter). 8. Forecasting of data forecasting of crime (Yhat) in form of cross sectional, time series or pooled data after examining of significance (discussion in next chapter). 9. Using the model for control or policy purpose (discussion in next chapter). 5. CAUSALITY AND CITREOUS PARIBUS OR UNDEFINED VARIABLES When one variable has a causal effect (causal means underlying or fundamental effect) on another variable; for example education is a variable has effect on HDI level, per capita earning

level, GDP of country, quality of work etc., here education has underlying or fundamental effect on PCI, HDI, GDP, Worker quality. Simply, casual effect expresses an association between two or more variables. For example price has effect on demand or price has effect on supply or price has effect on income or fertilizer has effect on crop yield etc., all these are examples of causal effect. The notion of citreous paribus, which means other (relevant) factors remain constant or remain equal or remain zero (0). Citreous paribus has an important role in causality analysis. For example price has effect on demand quantity citreous paribus (other factors remain constant or zero or fixed); such factors are income, substitute goods, complementary goods, advertising, taste, preference, etc. To determine the causal effect of Error Term or Other Factors or Residual or Citreous Paribus or Unobserved Variables is one of the most important objectives of econometric analysis; through econometric analysis we calculate the value of Citreous Paribus and finally estimate the value of dependent variable. Mathematically, Partial Function with intercept and slop y = 0 + 1x1 + 2x2 Partial Derivatives (a) y w.r.t. x1 (holding x2 constant or x2 treated as ceteris paribus or x2=0 ) (b) y w.r.t. x2 (holding x1 constant or x1 treated as ceteris paribus or x1=0 ) Here, Y = dependent, 0 = intercept, 1 is the slop of x1 and 2 is the slop of x2 From above equation ( c ), the changes in y, for given changes in x1 and x2 Or y for given x1 and x2 is y = 1x1 + 2x2 If x2 remain constant or x2 = 0, then we have y = 1x1,if x2 = 0 Where - 1 is the slop of the relationship of X1,

In other words, 1 is often called the partial effect of x1 on y. 1 = y / x1, if x2 = 0 If x1 remain constant or x1 = 0, then we have y = 1x2, if x1 = 0 Here 2 is the slop of the relationship of X2, In other words 2 is often called the partial effect of x2 on y. 2 = y / x2, if x1 = 0

8. ABOUT ECONOMETRIC SOFTWARE : EWIEWS/RATS/STATA/SAZAM/GRETL/SPSS.

Note : Student refer chapter 1 & 2 of handout.