Data analysis on Berkeley Guidance Study data

Data analysis on Berkeley Guidance Study data - 1 1 2345 67...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 1 1 2345 67 Data analysis on Berkeley Guidance Study data From the scatter plot matrix for the Berkeley Guidance Study data, we see that overall the somatotype is weakly related to each of the weight and height variables and there are a lot of variations. However, there are quite positive pair wise relationships between the covariates. For instance, WT9 and LG9 appeared to have a strong positive linear relationship between each other. So does WT18 and LG18 and so on. These interrelationships among covariates may enhance the joint relationship between response and the explanatory variables. Boxplot for gender Scatter plot of Somatotype vs. all other covariates Soma 10 14 18 20 40 60 25 35 50 80 30 36 42 1 4 7 10 16 WT2 HT2 8595 20 50 WT9 HT9 125 150 25 40 LG9 ST9 20 80 50 90 WT18 HT18 160190 3038 LG18 1 3 5 7 85 95 125 145 20 60 120 160 190 100 200 100 250 ST18 First, we fit the data by using the model with all the covariates in the study and the summary statistics is as below. ================================================================================ ===== Linear model: Soma ~ Sex + WT2 + HT2 + WT9 + HT9 + LG9 + ST9 + WT18 + HT18 + LG18 + ST18) Residuals: Min 1Q Median 3Q Max -1.70220 -0.46909 0.03840 0.43615 2.49553 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 14.457730 2.837676 5.095 1.26e-06 *** Sex 0.677846 0.370587 1.829 0.0698 . WT2 -0.144764 0.063802 -2.269 0.0250 * HT2 -0.028894 0.031382 -0.921 0.3590 WT9 0.097260 0.040607 2.395 0.0181 * HT9 -0.032578 0.033221 -0.981 0.3287 LG9 -0.111569 0.098735 -1.130 0.2607 ST9 0.003096 0.006715 0.461 0.6456 WT18 0.084229 0.016570 5.083 1.33e-06 *** HT18 -0.039793 0.021445 -1.856 0.0659 . LG18 0.049571 0.065148 0.761 0.4482 ST18 -0.015230 0.003584 -4.250 4.17e-05 ***--- Signif. codes: 0 *** 0.001 ** 0.01 * 0.05 . 0.1 1 Residual standard error: 0.754 on 124 degrees of freedom Multiple R-squared: 0.7465, Adjusted R-squared: 0.724 F-statistic: 33.19 on 11 and 124 DF, p-value: < 2.2e-16 ================================================================================ ===== If sorted by the t-value for each variable in the order from high to low (see table blow), we can see that the variable of weight at age 18 (WT18) and age 18 strength (ST18) are the most statistically significant in relation with the response. The next are the weight variables at age 9 and 2. For instance, per kg increase in a persons weight at age 18 will lead to approximately 0.08 unit increase toward obese body type if all other conditions weight at age 18 will lead to approximately 0....
View Full Document

This note was uploaded on 01/15/2010 for the course MATH 423 taught by Professor Steele during the Spring '06 term at McGill.

Page1 / 10

Data analysis on Berkeley Guidance Study data - 1 1 2345 67...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online