ebook img

Lec 3: Model Adequacy Checking PDF

30 Pages·2011·0.77 MB·English
by  Ying Li
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Lec 3: Model Adequacy Checking

Lec 3: Model Adequacy Checking Ying Li November 16, 2011 YingLi Lec3:ModelAdequacyChecking Model validation Model validation is a very important step in the model building procedure. (one of the most overlooked) A high R2 value does not guarantee that the model fits the data well. Use of a model that does not fit the data well can not provide good answers to the underlying scientific questions. YingLi Lec3:ModelAdequacyChecking An interesting example: Ascombe dataset 1011 l l 89 l l l l l l l y1 789 l l l l l l y2 67 l l 6 l 5 l 5 l 4 4 l 3 l 4 6 8 10 12 14 4 6 8 10 12 14 x1 x2 l l 12 12 10 10 y3 l y4 ll 68 l l l l l l l l l 68 llllllll 4 6 8 10 12 14 8 10 12 14 16 18 x3 x4 YingLi Lec3:ModelAdequacyChecking 1011 l l 89 l l l l l l l y1 456789 l l l l l l Rl_sabql==u30a..r05e00=000l11.667 y2 34567 l l l l R_sabq==u30a..r05e00=00011.667 4 6 8 10 12 14 4 6 8 10 12 14 x1 x2 l l 12 a=3.0001 12 b=0.5001 10 R_square=0.667 10 y3 l y4 ll 68 l l l l l l l l l 68 llllllll R_sabq==u30a..05re00=00011.667 4 6 8 10 12 14 8 10 12 14 16 18 x3 x4 YingLi Lec3:ModelAdequacyChecking Main Tool: Graphical residual Analysis Different types of plots of residuals (histgram, Plot of residuals in time sequence, plot of residuals versus fitted values) provide information on the adequacy of different aspects of the model. Graphical methods have an advantage over numerical methods in model validation graphical methods: a broad range of complex aspects numerical methods: narrowly focused on a particular aspect(a number) YingLi Lec3:ModelAdequacyChecking Why use residuals? If the model fit to the data were correct, the residuals would approximate the random errors. If the residuals appear to behave as the assumptions of the error it suggests the model fit the data well. Otherwise the model fits the data poorly.(non-normality, dependency, heteroscedasticity). YingLi Lec3:ModelAdequacyChecking Two concepts Homogeneity (Homoscedasticity) : In statistics, a sequence or a vector of random variables is homoscedastic if all random variables in the sequence or vector have the same finite variance. This is also known as homogeneity of variance. Heteroscedasticity: a collection of random variables is heteroscedastic, or heteroscedastic, if there are sub-populations that have different variabilities than others YingLi Lec3:ModelAdequacyChecking The assumptions for ANOVA y = µ+τ +(cid:15) ij i ij Assumptions of the errors: (cid:15) are normally distributed ij (cid:15) are independent ij (cid:15) has mean zero and constant variance σ2. ij YingLi Lec3:ModelAdequacyChecking Check the normality Normal probability plot. YingLi Lec3:ModelAdequacyChecking Check the normality Normal probability plot. YingLi Lec3:ModelAdequacyChecking

Description:
Lec 3: Model Adequacy Checking. Ying Li. November 16, 2011. Ying Li. Lec 3: Model Adequacy Checking. Page 2. Model validation. Model validation is a very
See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.