




版權(quán)說明:本文檔由用戶提供并上傳,收益歸屬內(nèi)容提供方,若內(nèi)容存在侵權(quán),請進行舉報或認(rèn)領(lǐng)
文檔簡介
1、Applied Econometrics,William Greene Department of Economics Stern School of Business,Applied Econometrics,15. Generalized Regression Model,Generalized Regression Model,Setting: The classical linear model assumes that E = Var = 2I. That is, observations are uncorrelated and all are drawn from a distr
2、ibution with the same variance. The generalized regression (GR) model allows the variances to differ across observations and allows correlation across observations.,Implications,The assumption that Var = 2I is used to derive the result Varb = 2(XX)-1. If it is not true, then the use of s2(XX)-1 to e
3、stimate Varb is inappropriate. The assumption was used to derive most of our test statistics, so they must be revised as well. Least squares gives each observation a weight of 1/n. But, if the variances are not equal, then some observations are more informative than others. Least squares is based on
4、 simple sums, so the information that one observation might provide about another is never used.,GR Model,The generalized regression model: y = X + , E|X = 0, Var|X = 2. Regressors are well behaved. We consider some examples Trace = n. (This is a normalization with no content.) Leading Cases Simple
5、heteroscedasticity Autocorrelation Panel data and heterogeneity more generally.,Least Squares,Still unbiased. (Proof did not rely on ) For consistency, we need the true variance of b, Varb|X = E(b-)(b-)|X = (XX)-1 EXX (XX)-1 = 2 (XX)-1 XX (XX)-1 . Divide all 4 terms by n. If the middle one converges
6、 to a finite matrix of constants, we have the result, so we need to examine (1/n)XX = (1/n)ij ij xi xj. This will be another assumption of the model. Asymptotic normality? Easy for heteroscedasticity case, very difficult for autocorrelation case.,Robust Covariance Matrix,Robust estimation: How to es
7、timate Varb|X = 2 (XX)-1 XX (XX)-1 for the LS b? The distinction between estimating 2 an n by n matrix and estimating 2 XX = 2 ijij xi xj NOTE VVVIRs for modern applied econometrics. The White estimator Newey-West.,The White Estimator,Groupwise Heteroscedasticity,Regression of log of per capita gaso
8、line use on log of per capita income, gasoline price and number of cars per capita for 18 OECD countries for 19 years. The standard deviation varies by country. The “solution” is “weighted least squares.”,Countries are ordered by the standard deviation of their 19 residuals.,White Estimator,+-+-+-+-
9、+-+-+ |Variable| Coefficient | Standard Error |t-ratio |P|T|t| Mean of X| +-+-+-+-+-+-+ Constant| 2.39132562 .11693429 20.450 .0000 LINCOMEP| .88996166 .03580581 24.855 .0000 -6.13942544 LRPMG | -.89179791 .03031474 -29.418 .0000 -.52310321 LCARPCAP| -.76337275 .01860830 -41.023 .0000 -9.04180473 |
10、White heteroscedasticity robust covariance matrix | +-+ Constant| 2.39132562 .11794828 20.274 .0000 LINCOMEP| .88996166 .04429158 20.093 .0000 -6.13942544 LRPMG | -.89179791 .03890922 -22.920 .0000 -.52310321 LCARPCAP| -.76337275 .02152888 -35.458 .0000 -9.04180473,Newey-West Estimator,Autocorrelate
11、d Residuals,Newey-West Estimate,+-+-+-+-+-+-+ |Variable | Coefficient | Standard Error |t-ratio |P|T|t | Mean of X| +-+-+-+-+-+-+ Constant -7.42451805 1.00168909 -7.412 .0000 LP -.00421871 .04895003 -.086 .9317 1.29943314 LY 3.98923512 .42831889 9.314 .0000 2.26889042 LPNC -.03342281 .06509844 -.513
12、 .6101 4.38036654 LPUC .00305706 .03093474 .099 .9217 4.10544881 LPPT -.07075631 .03635218 -1.946 .0577 4.14194132 | Robust VC Newey-West, Periods = 10 | +-+-+-+-+-+-+ |Variable | Coefficient | Standard Error |t-ratio |P|T|t | Mean of X| +-+-+-+-+-+-+ Constant -7.42451805 .71583767 -10.372 .0000 LP
13、-.00421871 .04863460 -.087 .9313 1.29943314 LY 3.98923512 .33181168 12.023 .0000 2.26889042 LPNC -.03342281 .04741084 -.705 .4844 4.38036654 LPUC .00305706 .04548241 .067 .9467 4.10544881 LPPT -.07075631 .03482236 -2.032 .0480 4.14194132,Generalized Least Squares,A transformation of the model: P = -
14、1/2. PP = -1 Py = PX + P or y* = X* + *. Why? E*|X*= PE|X*P = PE|XP = 2PP = 2 -1/2 -1/2 = 2 0 = 2I,Generalized Least Squares,Aitken theorem. The Generalized Least Squares estimator, GLS. Py = PX + P or y* = X* + *. E*|X*= 2I Use ordinary least squares in the transformed model. Satisfies the Gauss Ma
15、rkov theorem. b* = (X*X*)-1X*y*,Generalized Least Squares,Efficient estimation of and, by implication, the inefficiency of least squares b. = (X*X*)-1X*y* = (XPPX)-1 XPPy = (X-1X)-1 X-1y b. is efficient, so by construction, b is not.,Asymptotics for GLS,Asymptotic distribution of GLS. (NOTE. We appl
16、y the full set of results of the classical model to the transformed model. Unbiasedness Consistency - “well behaved data” Asymptotic distribution Test statistics,Unbiasedness,Consistency,Asymptotic Normality,Asymptotic Normality (Cont.),Test Statistics (Assuming Known ),With known , apply all famili
17、ar results to the transformed model: With normality, t and F statistics apply to least squares based on Py and PX With asymptotic normality, use Wald statistics and the chi-squared distribution, still based on the transformed model.,Generalized (Weighted) Least SquaresHeteroscedasticity Case,Autocor
18、relation,t = t-1 + ut (First order autocorrelation. How does this come about?) Assume -1 1. Why? ut = nonautocorrelated white noise t = t-1 + ut (the autoregressive form) = (t-2 + ut-1) + ut = . (continue to substitute) = ut + ut-1 + 2ut-2 + 3ut-3 + . = (the moving average form) (Some observations a
19、bout modeling time series.),Autocorrelation,Autocovariances,Autocorrelation Matrix,Generalized Least Squares,The Autoregressive Transformation,Unknown ,The problem (of course), is unknown. For now, we will consider two methods of estimation: Two step, or feasible estimation. Estimate first, then do
20、GLS. Emphasize - same logic as White and Newey-West. We dont need to estimate . We need to find a matrix that behaves the same as (1/n)X-1X. Properties of the feasible GLS estimator Maximum likelihood estimation of , 2, and all at the same time. Joint estimation of all parameters. Fairly rare. Some
21、generalities We will examine two applications: Harveys model of heteroscedasticity and Beach-MacKinnon on the first order autocorrelation model,Specification, must be specified first. A full unrestricted contains n(n+1)/2 - 1 parameters. (Why minus 1? Remember, tr() = n, so one element is determined
22、.) is generally specified in terms of a few parameters. Thus, = () for some small parameter vector . It becomes a question of estimating . Examples:,Harveys Model of Heteroscedasticity,Vari | X = 2 exp(zi) Covi,j | X = 0 e.g.: zi = firm size e.g.: zi = a set of dummy variables (e.g., countries) (The
23、 groupwise heteroscedasticity model.) 2 = diagonal exp( + zi), = log(2),AR(1) Model of Autocorrelation,Two Step Estimation,The general result for estimation when is estimated. GLS uses X-1XX -1 y which converges in probability to . We seek a vector which converges to the same thing that this does. C
24、all it “Feasible GLS” or FGLS, based on X XX y The object is to find a set of parameters such that X XX y - X -1 XX -1 y 0,Feasible GLS,Two Step FGLS,VVIR (Theorem 8.5) To achieve full efficiency, we do not need an efficient estimate of the parameters in , only a consistent one. Why?,Harveys Model,E
25、xamine Harveys model once again. Methods of estimation: Two step FGLS: Use the least squares residuals to estimate , then use X()-1 X-1X()-1y to estimate . Full maximum likelihood estimation. Estimate all parameters simultaneously. A handy result due to Oberhofer and Kmenta - the “zig-zag” approach.
26、 Examine a model of groupwise heteroscedasticity.,Harveys Model for Groupwise Heteroscedasticity,Groupwise sample, yig, xig, N groups, each with Ng observations. Varig = g2 Let dig = 1 if observation i,g is in group j, 0 else. = group dummy variable. Varig = g2 exp(2d2 + GdG) Var1 = g2 , Var2 = g2 e
27、xp(2) and so on.,Estimating Variance Components,OLS is still consistent: Est.Var1 = e1e1/N1 estimates g2 Est.Var2 = e2e2/N2 estimates g2 exp(2) Estimator of 2 is ln(e2e2/N2)/(e1e1/N1) (1) Now use FGLS weighted least squares Recompute residuals using WLS slopes (2) Recompute variance estimators Itera
28、te to a solution between (1) and (2),Applied Econometrics,William Greene Department of Economics Stern School of Business,Applied Econometrics,16. Applications of the Generalized Regression Model,Two Step Estimation of the Generalized Regression Model,Use the Aitken (Generalized Least Squares - GLS)
29、 estimator with an estimate of 1. is parameterized by a few estimable parameters. Examples, the heteroscedastic model 2. Use least squares residuals to estimate the variance functions 3. Use the estimated in GLS - Feasible GLS, or FGLS,General Result for Estimation When Is Estimated,True GLS uses X
30、-1 XX -1 y which converges in probability to . We seek a vector which converges to the same thing that this does. Call it FGLS, based on X -1 XX -1 y,FGLS,Feasible GLS is based on finding an estimator which has the same properties as the true GLS. Example Vari = 2 Exp(zi). True GLS would regress y/
31、Exp(1/2)zi) on the same transformation of xi. With a consistent estimator of , say s,c, we do the same computation with our estimates. So long as plim s,c = , FGLS is as good as true GLS.,FGLS vs. Full GLS,VVIR To achieve full efficiency, we do not need an efficient estimate of the parameters in , o
32、nly a consistent one. Why?,Heteroscedasticity,Setting: The regression disturbances have unequal variances, but are still not correlated with each other: Classical regression with hetero-(different) scedastic (variance) disturbances. yi = xi + i, Ei = 0, Vari = 2 i, i 0. The classical model arises if
33、 i = 1. A normalization: i i = 1. Not a restriction, just a scaling that is absorbed into 2. A characterization of the heteroscedasticity: Well defined estimators and methods for testing hypotheses will be obtainable if the heteroscedasticity is “well behaved” in the sense that i / i i 0 as n . I.e.
34、, no single observation becomes dominant. (1/n)i i some stable constant. (Not a probability limit as such.),GR Model and Testing,Implications for conventional estimation technique and hypothesis testing: 1. b is still unbiased. Proof of unbiasedness did not rely on homoscedasticity 2. Consistent? We
35、 need the more general proof. Not difficult. 3. If plim b = , then plim s2 = 2 (with the normalization).,Inference Based on OLS,What of s2(XX)-1 ? Depends on XX - XX. If they are nearly the same, the OLS covariance matrix is OK. When will they be nearly the same? Relates to an interesting property o
36、f weighted averages. Suppose i is randomly drawn from a distribution with Ei = 1. Then, (1/n)i i xi2 Ex2, just like (1/n)i xi2. This is the crux of the discussion in your text.,Inference Based on OLS,VIR: For the heteroscedasticity to be substantive wrt estimation and inference by LS, the weights mu
37、st be correlated with xs and/or their squares. (Text, page 220.) More likely, the heteroscedasticity will be important. Then, b is inefficient. (Later) The White estimator. ROBUST estimation of the variance of b. Implication for testing hypotheses. We will use Wald tests. Why? (ROBUST TEST STATISTIC
38、S),Finding Heteroscedasticity,The central issue is whether E2 = 2i is related to the xs or their squares in the model. Suggests an obvious strategy. Use residuals to estimate disturbances and look for relationships between ei2 and xi and/or xi2. For example, regressions of squared residuals on xs an
39、d their squares.,Procedures,Whites general test: nR2 in the regression of ei2 on all unique xs, squares, and cross products. Chi-squaredP Breusch and Pagans Lagrange multiplier test. Regress ei2 /(ee/n) 1 on Z (may be X). Chi-squared. Is nR2 with degrees of freedom rank of Z. (Very elegant.) Others
40、described in text for other purposes. E.g., groupwise heteroscedasticity. Wald, LM, and LR tests all examine the dispersion of group specific least squares residual variances.,Estimation: WLS form of GLS,General result - mechanics of weighted least squares. Generalized least squares - efficient esti
41、mation. Assuming weights are known. Two step generalized least squares: Step 1: Use least squares, then the residuals to estimate the weights. Step 2: Weighted least squares using the estimated weights. We develop a proof based on our asymptotic theory for the asymptotic equivalence of the second st
42、ep to true GLS.,Autocorrelation,The analysis of “autocorrelation” in the narrow sense of correlation of the disturbances across time largely parallels the discussions weve already done for the GR model in general and for heteroscedasticity in particular. One difference is that the relatively crisp r
43、esults for the model of heteroscedasticity are replaced with relatively fuzzy, somewhat imprecise results here. The reason is that it is much more difficult to characterize meaningfully “well behaved” data in a time series context. Thus, for example, in contrast to the sharp result that produces the
44、 White robust estimator, the theory underlying the Newey-West robust estimator is somewhat ambiguous in its requirement of a bland statement about “how far one must go back in time until correlation becomes unimportant.”,The AR(1) Model,t = t-1 + ut, | 1. Emphasize, this characterizes the disturbanc
45、es, not the regressors. A general characerization of the mechanism producing - history + innovations Analysis of this model in particular. The mean and variance and autocovariance “Stationarity.” Some general comments about “time series analysis.” (Not the subject of this course). Implication: The f
46、orm of 2 Var vs. Varu. Other models for autocorrelation - less frequently used - AR(1) is the workhorse.,Building the Model,Prior view: A feature of the data “Account for autocorrelation in the data. Different models, different estimators Contemporary view: Why is there autocorrelation? What is miss
47、ing from the model Build in appropriate dynamic structures Autocorrelation should be “built out” of the model Use robust procedures (Newey-West) instead of elaborate models specifically for the autocorrelation.,Model Misspecification,Implications for Least Squares,Familiar results: Consistent, unbia
48、sed, inefficient, asymptotic normality The inefficiency of least squares: Difficult to characterize generally. It is worst in “l(fā)ow frequency” i.e., long period (year) slowly evolving data. Can be extremely bad. GLS vs. OLS, the efficiency ratios can be 3 or more. A very important exception - the lag
49、ged dependent variable yt = xt + yt-1 + t. t = t-1 + ut,. Obviously, Covyt-1 ,t 0, because of the form of t. How to estimate? IV Should the model be fit in this form? Something missing? Robust estimation of the covariance matrix - the Newey-West estimator.,GLS and FGLS,Theoretical result for known -
50、 i.e., known . Prais-Winsten vs. Cochrane-Orcutt. FGLS estimation: How to estimate ? OLS residuals as usual - first autocorrelation. Many variations, all based on correlation of et and et-1 a. Prais-Winsten vs. Cochrane-Orcutt. b. The question of dropping the first observation. Should you?,Testing f
51、or Autocorrelation,A general proposition: There are several tests. All are functions of the simple autocorrelation of the least squares residuals. The Durbin - Watson test. d 2(1 - r). Small values of d lead to rejection of NO AUTOCORRELATION: Why are the bounds necessary? Godfreys LM test. Regression of et on et-1 and xt. Uses a “partial correlation.” Durbins H test when lagged y is present. H = (1 - d/2) (T/(1 - T Est.Varc)1/2 where c is the coefficient on the lagged y. If it is not computable, use Godfreys test. (Durbin discovered it earlier.),Time Ser
溫馨提示
- 1. 本站所有資源如無特殊說明,都需要本地電腦安裝OFFICE2007和PDF閱讀器。圖紙軟件為CAD,CAXA,PROE,UG,SolidWorks等.壓縮文件請下載最新的WinRAR軟件解壓。
- 2. 本站的文檔不包含任何第三方提供的附件圖紙等,如果需要附件,請聯(lián)系上傳者。文件的所有權(quán)益歸上傳用戶所有。
- 3. 本站RAR壓縮包中若帶圖紙,網(wǎng)頁內(nèi)容里面會有圖紙預(yù)覽,若沒有圖紙預(yù)覽就沒有圖紙。
- 4. 未經(jīng)權(quán)益所有人同意不得將文件中的內(nèi)容挪作商業(yè)或盈利用途。
- 5. 人人文庫網(wǎng)僅提供信息存儲空間,僅對用戶上傳內(nèi)容的表現(xiàn)方式做保護處理,對用戶上傳分享的文檔內(nèi)容本身不做任何修改或編輯,并不能對任何下載內(nèi)容負(fù)責(zé)。
- 6. 下載文件中如有侵權(quán)或不適當(dāng)內(nèi)容,請與我們聯(lián)系,我們立即糾正。
- 7. 本站不保證下載資源的準(zhǔn)確性、安全性和完整性, 同時也不承擔(dān)用戶因使用這些下載資源對自己和他人造成任何形式的傷害或損失。
最新文檔
- 2025年六路一體化硅箱項目可行性研究報告
- -模塊化-橇裝化-工廠化-集成技術(shù)在蘇丹石油地面工程建設(shè)中的應(yīng)用
- 2025年螺旋輸送網(wǎng)帶項目可行性研究報告
- 淺談施工單位“安管人員”安全生產(chǎn)責(zé)任
- 2024-2025學(xué)年高中歷史專題六西方人文精神的起源與發(fā)展二神權(quán)下的自我課后篇鞏固探究人民版必修3
- 淺談建筑施工企業(yè)工程項目成本管理
- 心理素質(zhì)拓展心得體會范文(30篇)
- Unit 1 Life Choices Lesson 2 Understanding and Coping with Stress 教學(xué)設(shè)計-2024-2025學(xué)年高中英語北師大版(2019)必修第一冊
- 2023-2029年中國異步發(fā)電機行業(yè)市場全景評估及投資前景展望報告
- 2025-2030年中國拖拉機加長半軸項目投資可行性研究分析報告
- 黔2022-T122 磷石膏砂漿噴筑復(fù)合墻標(biāo)準(zhǔn)圖集 第1部分:輕鋼龍骨-磷石膏砂漿噴筑復(fù)合墻體
- GB 29444-2012煤炭井工開采單位產(chǎn)品能源消耗限額
- 精細(xì)化學(xué)品化學(xué)-緒論課件
- 車間維修電工安全技術(shù)操作規(guī)程
- 第四章 聚合物基納米復(fù)合材料課件
- 教學(xué)課件 211和985工程大學(xué)簡介
- 臥式水泵安裝
- 綜合交通運輸體系課件
- 趣味經(jīng)濟學(xué)課件
- 實木家具生產(chǎn)標(biāo)準(zhǔn)工藝標(biāo)準(zhǔn)流程
- 小學(xué)生讀書卡模板
評論
0/150
提交評論