If you have access to data on an entire population, say the size of every house in Ames, Iowa, it’s straight forward to answer questions like, “How big is the typical house in Ames?” and “How much variation is there in sizes of houses?”. If you have access to only a sample of the population, as is often the case, the task becomes more complicated. What is your best guess for the typical size if you only know the sizes of several dozen houses? This sort of situation requires that you use your sample to make inference on what your population looks like.
In the previous lab, ``Sampling Distributions’’, we looked at the population data of houses from Ames, Iowa. Let’s start by loading that data set.
load("more/ames.RData")
In this lab we’ll start with a simple random sample of size 60 from the population. Specifically, this is a simple random sample of size 60. Note that the data set has information on many housing variables, but for the first portion of the lab we’ll focus on the size of the house, represented by the variable Gr.Liv.Area
.
population <- ames$Gr.Liv.Area
samp <- sample(population, 60)
The distribution is right skewed. The typical size within my sample is somewhere around 1387 - I am interpreting “typical” as Mean.
summary(samp)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 698 1236 1398 1467 1637 2668
hist(samp)
I don’t expect another student’s distribution to be identical to mine - 60 is not a large sample so a lot of variability is possible. I would be somewhat similar but could be skewed left instead of right and spread and mean vary.
One of the most common ways to describe the typical or central value of a distribution is to use the mean. In this case we can calculate the mean of the sample using,
sample_mean <- mean(samp)
Return for a moment to the question that first motivated this lab: based on this sample, what can we infer about the population? Based only on this single sample, the best estimate of the average living area of houses sold in Ames would be the sample mean, usually denoted as \(\bar{x}\) (here we’re calling it sample_mean
). That serves as a good point estimate but it would be useful to also communicate how uncertain we are of that estimate. This can be captured by using a confidence interval.
We can calculate a 95% confidence interval for a sample mean by adding and subtracting 1.96 standard errors to the point estimate (See Section 4.2.3 if you are unfamiliar with this formula).
se <- sd(samp) / sqrt(60)
lower <- sample_mean - 1.96 * se
upper <- sample_mean + 1.96 * se
c(lower, upper)
## [1] 1364.615 1569.885
This is an important inference that we’ve just made: even though we don’t know what the full population looks like, we’re 95% confident that the true average size of houses in Ames lies between the values lower and upper. There are a few conditions that must be met for this interval to be valid.
The data must be sampled randomly.The sample values must be independent of each other. The sample size must be sufficiently large (n>30).
Suppose we took many samples and built a confidence interval from each sample, then about 95% of those intervals would contain the actual mean, μ.
In this case we have the luxury of knowing the true population mean since we have data on the entire population. This value can be calculated using the following command:
mean(population)
## [1] 1499.69
Yes. My interval was (1261, 1513).
95% of those intervals should capture our true mean since that is the definition of 95% confidence interval.
Using R, we’re going to recreate many samples to learn more about how sample means and confidence intervals vary from one sample to another. Loops come in handy here (If you are unfamiliar with loops, review the Sampling Distribution Lab).
Here is the rough outline:
But before we do all of this, we need to first create empty vectors where we can save the means and standard deviations that will be calculated from each sample. And while we’re at it, let’s also store the desired sample size as n
.
samp_mean <- rep(NA, 50)
samp_sd <- rep(NA, 50)
n <- 60
Now we’re ready for the loop where we calculate the means and standard deviations of 50 random samples.
for(i in 1:50){
samp <- sample(population, n) # obtain a sample of size n = 60 from the population
samp_mean[i] <- mean(samp) # save sample mean in ith element of samp_mean
samp_sd[i] <- sd(samp) # save sample sd in ith element of samp_sd
}
Lastly, we construct the confidence intervals.
lower_vector <- samp_mean - 1.96 * samp_sd / sqrt(n)
upper_vector <- samp_mean + 1.96 * samp_sd / sqrt(n)
Lower bounds of these 50 confidence intervals are stored in lower_vector
, and the upper bounds are in upper_vector
. Let’s view the first interval.
c(lower_vector[1], upper_vector[1])
## [1] 1346.275 1684.559
Using the following function (which was downloaded with the data set), plot all intervals. What proportion of your confidence intervals include the true population mean? Is this proportion exactly equal to the confidence level? If not, explain why.
lower_vector
## [1] 1346.275 1378.720 1307.593 1411.046 1398.546 1367.169 1345.872
## [8] 1405.389 1377.343 1419.418 1245.942 1494.451 1314.968 1389.131
## [15] 1414.166 1430.506 1350.828 1263.393 1362.897 1456.028 1420.914
## [22] 1460.747 1331.375 1288.673 1423.053 1285.051 1426.613 1300.339
## [29] 1360.892 1360.666 1435.377 1367.397 1449.293 1453.905 1313.147
## [36] 1247.880 1325.097 1381.811 1329.695 1374.951 1263.463 1308.995
## [43] 1457.273 1413.533 1447.599 1359.420 1308.470 1363.907 1286.009
## [50] 1334.826
upper_vector
## [1] 1684.559 1616.846 1568.040 1694.954 1648.254 1623.097 1581.095
## [8] 1635.611 1605.090 1681.982 1472.791 1726.216 1554.365 1743.569
## [15] 1652.301 1697.828 1591.339 1559.274 1588.270 1689.538 1640.452
## [22] 1730.053 1551.191 1536.327 1640.180 1529.249 1692.187 1534.661
## [29] 1602.341 1610.801 1681.656 1614.236 1685.474 1702.428 1528.920
## [36] 1439.353 1560.003 1609.656 1544.072 1618.849 1514.203 1552.038
## [43] 1767.727 1655.834 1775.901 1616.080 1516.096 1605.327 1471.991
## [50] 1631.641
plot_ci(lower_vector, upper_vector, mean(population))
Only 3 out of 50 intervals don’t include the true mean. So 47/50 include the true mean:
47/50*100
## [1] 94
94% include the mean. It does not exactly equal 95% because this is still an estimate - but is very close!
Pick a confidence level of your choosing, provided it is not 95%. What is the appropriate critical value?
92%, critical value is:
qnorm(0.04, lower.tail=FALSE)
## [1] 1.750686
plot_ci
function, plot all intervals and calculate the proportion of intervals that include the true population mean. How does this percentage compare to the confidence level selected for the intervals?lower_vector <- samp_mean - 1.75 * samp_sd / sqrt(n)
upper_vector <- samp_mean + 1.75 * samp_sd / sqrt(n)
lower_vector
## [1] 1364.397 1391.477 1321.546 1426.256 1411.923 1380.880 1358.473
## [8] 1417.722 1389.544 1433.484 1258.095 1506.867 1327.793 1408.118
## [15] 1426.923 1444.827 1363.712 1279.244 1374.970 1468.538 1432.675
## [22] 1475.174 1343.151 1301.940 1434.685 1298.133 1440.840 1312.892
## [29] 1373.827 1374.066 1448.571 1380.621 1461.946 1467.219 1324.706
## [36] 1258.138 1337.681 1394.017 1341.180 1388.017 1276.896 1322.016
## [43] 1473.904 1426.513 1465.186 1373.170 1319.593 1376.840 1295.972
## [50] 1350.726
upper_vector
## [1] 1666.436 1604.090 1554.088 1679.744 1634.877 1609.387 1568.493
## [8] 1623.278 1592.889 1667.916 1460.639 1713.800 1541.541 1724.582
## [15] 1639.543 1683.507 1578.455 1543.423 1576.196 1677.029 1628.691
## [22] 1715.626 1539.416 1523.060 1628.549 1516.167 1677.960 1522.108
## [29] 1589.406 1597.401 1668.463 1601.013 1672.821 1689.114 1517.360
## [36] 1429.095 1547.419 1597.450 1532.587 1605.783 1500.771 1539.018
## [43] 1751.096 1642.854 1758.314 1602.330 1504.974 1592.394 1462.028
## [50] 1615.740
plot_ci(lower_vector, upper_vector, mean(population))
#Proportion of intervals that include true mean:
46/50*100
## [1] 92
This is a product of OpenIntro that is released under a Creative Commons Attribution-ShareAlike 3.0 Unported. This lab was written for OpenIntro by Andrew Bray and Mine Çetinkaya-Rundel.