From Factor analysis to SEM!

Notes

MSc stats notes

https://github.com/tbates/Multivariate-Stats-Course

http://tbates.github.io/Multivariate-Stats-Course

https://cran.r-project.org/web/packages/psych/vignettes/psych_for_sem.pdf

library(arm)
m1 = lm(mpg ~ wt * disp, data = mtcars)
arm::display(arm::standardize(m1))

require(umx)
data(demoOneFactor)
latents   = c("g")
manifests = names(demoOneFactor)
myData    = mxData(cov(demoOneFactor), type = "cov", numObs = 500)
m1 <- umxRAM("OneFactor", data = myData,
	umxPath(latents, to = manifests),
	umxPath(var = manifests),
	umxPath(var = latents, fixedAt = 1)
)

umxSummary(m1, showEstimates = "std")
plot(m1)

Openness effect on politics is mediated by NFC.

Transforms

http://stats.stackexchange.com/questions/298/in-linear-regression-when-is-it-appropriate-to-use-the-log-of-an-independent-va/3530#3530

Transform the IV

http://stats.stackexchange.com/questions/35711/box-cox-like-transformation-for-independent-variables/35717#35717

Fit

Why AIC and not R^2?

http://stats.stackexchange.com/questions/13314/is-r2-useful-or-dangerous/13317#13317

Binary and ordinal outcomes

http://stats.stackexchange.com/questions/20523/difference-between-logit-and-probit-models

Classic datasets

Iris_flower_data_set

Approaches

scikit sklearn TODO: Get google ML tutorial from youtube history 2016-04-24 04:10AM

http://onlinelibrary.wiley.com/doi/10.1111/pops.12318/abstract There is growing interest in how genes affect political beliefs. To better understand the role of genes in politics, we examine the relationship between cognitive style (the need for cognition, the need for cognitive closure) and various measures of political attitudes (issue-based ideology, identity-based ideology, social ideology, economic ideology, authoritarianism, and egalitarianism). We show, for the first time, that the need for cognition and the need for cognitive closure are heritable and are linked to political ideology primarily, perhaps solely, because of shared genetic influences; these links are stronger for social than economic ideology. Although prior research demonstrated that Openness to Experience shares genetic variance with political ideology, we find that these measures of cognitive style account for distinct genetic variance in political ideology. Moreover, the genetic Openness-ideology link is fully accounted for by the need for cognition. This combination of findings provides a clearer understanding of the role of genes in political beliefs and suggests new directions for research on Big Five personality traits and ideology.

Meta-analysis

Systematic reviews can be dangerous http://onlinelibrary.wiley.com/doi/10.1111/1468-0009.12210/full

Cars