Advance statistics 2
Upcoming SlideShare
Loading in...5
×
 

Advance statistics 2

on

  • 1,071 views

 

Statistics

Views

Total Views
1,071
Slideshare-icon Views on SlideShare
1,071
Embed Views
0

Actions

Likes
1
Downloads
6
Comments
0

0 Embeds 0

No embeds

Accessibility

Upload Details

Uploaded via as Microsoft PowerPoint

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    Advance statistics 2 Advance statistics 2 Presentation Transcript

    • Advance Statistics: Introductionto Regressionand Analysis of VarianceFixed and Random Effects
    • Today’s Report Random effects. One-way random effects ANOVA. Two-way mixed & random effects ANOVA. Sattherwaite’s procedure.
    • Two-way ANOVA Second generalization: more than one grouping variable. Two-way ANOVA model: observations: (Yijk), 1 < i < r , 1 < j < m, 1 < k < nij : r groups in first grouping variable, m groups ins second and nij samples in (i; j)-“cell”: Yijk = µ + α i + βj + (α β)ij + Eijk, Eijk~N(0, σ ² ): n Constraints: r∑i=1 αi = 0 m∑j=1 βj = 0 m∑j=1 (αβ)ij = 0, 1 < i < r r∑i=1 (αβ)ij = 0, 1 < j < m
    • Random and fixed effects In ANOVA examples we have seen so far, the categorical variables are well-defined categories: below average fitness, long duration, etc. In some designs, the categorical variable is “subject”. Simplest example: repeated measures, where more than one (identical) measurement is taken on the same individual. In this case, the “group” effect αi is best thought of as random because we only sample a subset of the entire population of subjects.
    • When to use random effects? A “group” effect is random if we can think of the levels we observe in that group to be samples from a larger population. Example: if collecting data from different medical centers, “center” might be thought of as random. Example: if surveying students on different campuses, “campus” may be a random effect.
    • Example: sodium content in beer How much sodium is there in North American beer? How? much does this vary by brand? Observations: for 6 brands of beer, researchers recorded the sodium content of 8 12 ounce bottles. Questions of interest: what is the “grand mean” sodium content? How much variability is there from brand to brand? “Individuals” in this case are brands, repeated measures are the 8 bottles.
    • One-way random effects model Suppose we take n identical measurements from r subjects Yij ~ µ + αi + Eij, 1 < I < r, 1 < j < n Eij ~ N(0, σ²), 1 < i < r, 1 < j <n αi ~ N(0,σ²µ), 1 < i < r We might be interested in the population mean, µ. : CIs, is it zero? etc. Alternatively, we might be interested in the variability across subjects,σ²µ : CIs, is it zero?
    • One-way random ANOVA table Source SS df E(MS) Treatments SSTR = r ∑i=1 n(Y1 – Y..) ² r–1 σ² + n σ²µ Error SSE = r ∑i=1 (Yij- Yi.) ² (n - 1)r σ² Only change here is the expectation of SSTR which reflects randomness of αi’s ANOVA table is still useful to setup tests: the same F statistics for fixed or random will work here. Under Ho : σ²µ = 0, it is to see that MSTR ~Fr-1, (n-1) r’ MSE
    • Inference for µ We know that E(Y..) = µ. , and can show that Var(Y..) = nσ²µ + σ² rn Therefore, Y.. - µ square SSTR ~tr - 1 (r-1)rn Why r – 1 Degrees of freedom? Imagine we could record an infinite number of observations for each individual, so that Yi ---- µi To learn anything about µ. We still only have r observations (µ1,…., µr). Sampling more within an individual cannot narrow the CI for µ,.
    • Estimating ²σµ From the ANOVA table σ²µ = E(SSTR/(r-1))-E(SSE/((n-1)r)) n Natural estimate S²µ = SSTR/(r-1)-SSE/((n-1)r)) n Problem: this estimate can be negative! One of the difficulties in random effect model
    • Example: productivity study Imagine a study on the productivity of employees in a large manufacturing company. Company wants to get an idea of daily productivity, and how it depends on which machine on employee uses. Study: take m employee and r machines, having each employee work on each machine for a total of n days. As these employees are not all employees, and these machines are not all machines it take sense to think of both the effect of machine and employees (and interactions) as random
    • Two-way random effects model Yijk ~ µ.. + αi + βj + (αβ)ij + Eij, 1< i < r, 1 < j < m, 1 < k < n Eijk ~ N(0, σ²), 1 < i < r, 1 < j < m, 1 < k < n αi ~ N(0, σ²α), 1 < i < r. βj ~ N(0, σ²β), 1 < j < m. (αβ)ij ~ N(0, ²σαβ), 1 < j < m, 1< i < r. Cov (Yijk,Yi’j’k’) = &ii’ ²σα + &jj’ ²σβ + &ii’ &jj’ ²σαβ + &ii’ &jj’ &kk’ σ².
    • ANOVA table: Two-way (random)SS df E(SS)SSA = nm r∑i=1 (Yi.. – Y…) ² r -1 σ² + nm σ²α + nσ²αβSSB = nr m∑j=1 (Yj.. – Y…) ² m -1 σ² +nr σ²β + nσ²αβSSAB = n r∑i=1 (Yij. – Yi.. –Yj.. + Y…) ² (m – 1)(r – 1) σ² + nσ²αβSSE = r∑i=1 m∑j=1 n ∑k=1 (Yijk – Y ij.)² (n – 1)ab σ² To test Ho : σ²α = 0 use SSA and SSAB. To test H0 : σ²αβ = 0 use SSAB and SSE.
    • Mixed effects model In some studies, some factors can be thought of as fixed, others random. For instance, we might have a study of the effect of a standard part of the brewing process on sodium levels in the beer example Then, we might think of a model in which we have a fixed effect for “brewing technique” and a random effect for beer
    • Two-way mixed effects model Yijk ~ µ.. + αi + βj + (αβ)ij + Eij, 1< i < r, 1 < j < m, 1 < k < n Eijk ~ N(0, σ²), 1 < i < r, 1 < j < m, 1 < k < n αi ~ N(0, σ²α), 1 < i < r. βj ~ N(0, σ²β), 1 < j < m. are constant! (αβ)ij ~ N (0,(m-1) σ²αβ/m), 1 < j < m, 1< i < r. Constraints: m∑j=1 βj = 0 m∑j=1 (αβ)ij = 0, 1 < i < r Cov ((αβ)ij, (αβ) i’j’ ) = - σ²αβ/m Cov (Yijk,Yi’j’k’) = &jj’ (σ²β + &ii’ m-1 σ²β – (1-&ii’) 1 σ²αβ + &ii’ &kk’ σ²). m m
    • ANOVA table: Two-way (mixed)SS df E(MS)SSA r -1 σ² + nm σ²αSSB m -1 σ² + m∑j=1 β²i+ nσ²αβ m -1SSAB (m – 1)(r – 1) σ² + nσ²αβSSE = r∑i=1 m∑j=1 n ∑k=1 (Yijk – Y ij.)² (n – 1)ab σ² To test Ho : σ²α = 0 use SSA and SSAB. To test Ho : β1 = … = βm = 0 0 use SSB and SSAB. To test Ho : σ²αβ use SSAB and SSE.
    • Confidence intervals for variances Consider estimating σ²β in two-way random effects ANOVA. A natural estimate is σ²β = nr (MSB – MSAB) What about CI? A linear combination of x² - but not x². To form a confidence interval for σ²β we need to know distribution of a linear combination of MS.’s at least approximately.
    • Satterwaite’s procedure Given K independent MS.’s L ~ k∑i=1 ciMSi Then dfTL “~” x²dfT E(L) where dfT = (k∑i=1 ciMSi)² k∑i=1 ciMS²i/dfi Where dfi the the degrees of freedom of the i-th MS. (1-α) . 100% CI for E(L): LL = dfT x L , LU = dfT x L x² dfT; 1-α/2 x² dfT; 1-α/2