After our Intro to Stan lecture I think it would be valuable to have you go through a similar exercise. Let’s test a second research question.
Research question: Is sea ice extent declining in the Southern Hemisphere over time? Is the same pattern happening in the Antarctic as in the Arctic? Fit a Stan model to find out!
Make sure you follow the steps we used in class.
What do your Stan model results indicate so far?
#place the code here
seaice <- read.csv("C:/Users/sharo/Desktop/seaice.csv", stringsAsFactors=F)
colnames(seaice) <- c("year", "extent_north", "extent_south")
head(seaice)
## year extent_north extent_south
## 1 1979 12.328 11.700
## 2 1980 12.337 11.230
## 3 1981 12.127 11.435
## 4 1982 12.447 11.640
## 5 1983 12.332 11.389
## 6 1984 11.910 11.454
#plot data
plot(extent_south ~ year, pch = 20, data = seaice)
#write the code
lm1 <- lm(extent_south ~ year, data = seaice)
summary(lm1)
##
## Call:
## lm(formula = extent_south ~ year, data = seaice)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.23372 -0.18142 0.01587 0.18465 0.88814
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -14.199551 10.925576 -1.300 0.2018
## year 0.012953 0.005468 2.369 0.0232 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3843 on 37 degrees of freedom
## Multiple R-squared: 0.1317, Adjusted R-squared: 0.1082
## F-statistic: 5.611 on 1 and 37 DF, p-value: 0.02318
#write the code here
x <- I(seaice$year - 1978)
y <- seaice$extent_south
N <- length(seaice$year)
lm1 <- lm(y ~ x)
summary(lm1)
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.23372 -0.18142 0.01587 0.18465 0.88814
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 11.421555 0.125490 91.015 <2e-16 ***
## x 0.012953 0.005468 2.369 0.0232 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3843 on 37 degrees of freedom
## Multiple R-squared: 0.1317, Adjusted R-squared: 0.1082
## F-statistic: 5.611 on 1 and 37 DF, p-value: 0.02318
lm_alpha <- summary(lm1)$coeff[1]; lm_alpha
## [1] 11.42155
lm_beta <- summary(lm1)$coeff[2]; lm_beta
## [1] 0.01295304
lm_sigma <- sigma(lm1); lm_sigma
## [1] 0.384331
#write the code
stan_data <- list(N = N, x = x, y = y)
write("// Stan model for simple linear regression
data {
int < lower = 1 > N; // Sample size
vector[N] x; // Predictor
vector[N] y; // Outcome
}
parameters {
real alpha; // Intercept
real beta; // Slope (regression coefficients)
real < lower = 0 > sigma; // Error SD
}
model {
y ~ normal(alpha + x * beta , sigma);
}
generated quantities {
} // The posterior predictive distribution",
"stan_model1.stan")
#Let's save that file path
stan_model1 <- "stan_model1.stan"
#code here
fit <- stan(file = stan_model1, data = stan_data, warmup = 500, iter = 1000, chains = 4, cores = 2, thin = 1)
fit
## Inference for Stan model: stan_model1.
## 4 chains, each with iter=1000; warmup=500; thin=1;
## post-warmup draws per chain=500, total post-warmup draws=2000.
##
## mean se_mean sd 2.5% 25% 50% 75% 97.5% n_eff Rhat
## alpha 11.42 0.00 0.13 11.14 11.33 11.42 11.50 11.68 760 1
## beta 0.01 0.00 0.01 0.00 0.01 0.01 0.02 0.02 786 1
## sigma 0.40 0.00 0.05 0.32 0.37 0.39 0.43 0.50 808 1
## lp__ 16.27 0.05 1.24 13.04 15.73 16.56 17.20 17.71 757 1
##
## Samples were drawn using NUTS(diag_e) at Tue Mar 24 19:34:14 2020.
## For each parameter, n_eff is a crude measure of effective sample size,
## and Rhat is the potential scale reduction factor on split chains (at
## convergence, Rhat=1).
#code here
posterior <- extract(fit)
str(posterior)
## List of 4
## $ alpha: num [1:2000(1d)] 11.4 11.7 11.4 11.2 11.4 ...
## ..- attr(*, "dimnames")=List of 1
## .. ..$ iterations: NULL
## $ beta : num [1:2000(1d)] 0.01072 0.00668 0.01405 0.01875 0.01576 ...
## ..- attr(*, "dimnames")=List of 1
## .. ..$ iterations: NULL
## $ sigma: num [1:2000(1d)] 0.458 0.468 0.403 0.384 0.424 ...
## ..- attr(*, "dimnames")=List of 1
## .. ..$ iterations: NULL
## $ lp__ : num [1:2000(1d)] 16.4 14.7 17.7 16.1 17.3 ...
## ..- attr(*, "dimnames")=List of 1
## .. ..$ iterations: NULL
#code here
plot(y ~ x, pch = 20, main="Comaring the resut of Linear and Stan Model")
abline(lm1, col = "green", pch=22, lty = 2, lw = 3)
abline( mean(posterior$alpha), mean(posterior$beta), col = "yellow", lw = 1)
legend("topleft",c("Linear Model","Stan Model"),fill=c("green","yellow"))
#The results are identical to our “lm” model.
#code here
plot(y ~ x, pch = 20, main="Comaring the result of Linear and 'All' Stan Models")
for (i in 1:500) {
abline(posterior$alpha[i], posterior$beta[i], col = "blue", lty = 1)
}
abline(lm1, col = "green", pch=22, lty = 2, lw = 3)
abline( mean(posterior$alpha), mean(posterior$beta), col = "yellow", lw = 1)
legend("topleft",c("Linear Model","Mean Stan Model", "All stan Models"),fill=c("green","yellow","blue"))