一、异方差的概念及类型 二、产生异方差的背景 三、实际经济问题中的异方差性 四、异方差性的后果 五、异方差性的检验 六、异方差的修正 七、案例

▪ 10.1 时间序列分解 ▪ 10.2 长期趋势分析 ▪ 10.3 季节变动分析 ▪ 10.4 循环波动分析 ▪ 10.5 时间序列的自相关分析 ▪ 10.6 时间序列的动态分析指标 ▪ 10.7 景气循环分析

12.1 The Nature of Autocorrelation 1. Definition (1) CLRM assumption: No autocorrelation exist in dishurbances ui; E(iμi)=0 Autocorrelation means: E(μiμ)≠0 (2) Autocorrelation is usually associated with time series data, but it can also occur in cross-sectional data, which is called spatial correlation

The models we discussed are models that are linear in parameters; variables Y and Xs do not necessarily have to be linear The price elasticity of demand~the log-linear models The rate of growth~semilog model Functional forms of regression models which are linear in parameters, but not necessarily linear in variables:

Dummy variables a dummy variable is a variable that takes on the value l or o Examples: male(= 1 if are male, O otherwise), south(=l if in the south, 0 otherwise), etc dummy variables are also called binar variables. for obvious reasons Economics 20- Prof anderson

The Normal Distribution: the distribution of a continuous r.v. whose value depends on a number of factors, yet no single factor dominates the other. 1. Properties of the normal distribution: 1)The normal distribution curve is symmetrical around its mean valueu. 2)The PDF of the distribution is the highest at its mean value but tails off at its extremities

Definition (1) Econometrics: economic measurement. (2) Econometrics: the social science in which the tools of economic theory, mathematics, and statistical inference are applied to the analysis of economic phenomena. (3) Econometrics: the result of a certain outlook on the role of economics, consists of the application of mathematical statistics to economic data to lend

What is Heteroskedasticity Recall the assumption of homoskedastic implied that conditional on the explanator variables the variance of the unobserved error u was constant If this is not true that is if the variance of u is different for different values of thex's. then the errors are heteroskedastic

Functional form e We' ve seen that a linear regression can really fit nonlinear relationships 2 Can use logs on RHS, LHS or both Can use quadratic forms ofx's Can use interactions ofx's e How do we know if we've gotten the right functional form for our model? Economics 20- Prof anderson

Stationary Stochastic Process e A stochastic process is stationary if for every collection of time indices 11 e Thus, stationarity implies that the x,'s are dentically distributed and that the nature of any correlation between adjacent terms is