Estimation

• Objectives and Introduction

• Confidence Interval for 
• Interpretation of Confidence Intervals
• Conf. Intervals for Continuous Data in JMP
• Confidence Interval for a Proportion
• Confidence Interval for a Proportion in JMP
• Confidence Interval for a Poisson Mean
• Sample Size Determination
• Conclusion

1
 Objectives
After completing this section, participants should be able to:
 Interpret confidence intervals for a mean or for a
proportion.
 Be able to obtain confidence intervals for a mean or
proportion using JMP.
 Understand the impact on a confidence interval of
changing the sample size or the confidence level.
 Understand that they can compute how large a sample is
needed to obtain a desired precision.
 Understand the key terms: confidence level, confidence
interval, standard error, bound on error.
2
 Introduction
The populations of occurrences or items associated with industrial
processes are almost always conceptual, in the sense that the
process can produce infinitely many occurrences or parts.
This means that, when we obtain data from these processes, we are
obtaining data on a sample from the underlying population.
Our main interest is focused on the parameters (summary
measures) of the population, and not the sample statistics.
For this reason, we need to address the topic of using the sample to
draw conclusions, or inferences, about the parameters of the
underlying population.
This topic is known as inferential statistics.

3
 Introduction
Suppose that we are interested in the time spent in resolving a
certain type of incident, for example, a customer complaint.
For a sample of observed incidents, we can calculate the sample
mean.
The sample mean is a point estimate of the population mean,
namely, it is an estimate of the mean time for resolving all
incidents that might have occurred.
Likewise, the sample variance of resolution times is an estimate
of the population variance.
Unfortunately, the true underlying parameter value may be
far from the estimate obtained from a sample.

4
 Introduction
The mean of the population, for instance, may fall far from the
sample mean of a random sample from that population.
What is needed is a range, or interval, of values that is likely to
contain the value of the parameter of interest.
Such intervals are called confidence intervals.
Intuitively, a confidence interval is a range of plausible values
for the population parameter.
The width of the confidence interval is important, as it indicates
how precisely we can estimate the parameter of interest.
Narrow intervals are desirable, since they give a narrow range of
values likely to contain the parameter of interest, meaning that
our knowledge about the parameter value is precise.
5
 Confidence Interval for 
We begin by developing the idea of a confidence interval for the
mean of a continuous characteristic.
For a continuous distribution, we want to calculate an interval of
values likely to contain the true unknown population mean .
Example: Suppose that we are interested in staff hours spent in
resolving a given type of incident or complaint.
A random sample of 20 such incidents is identified, and the time
in hours to resolve each of these is determined.
We are interested in the mean resolution time for all such
incidents, in other words, we are interested in , the mean of
the population of resolution times.
6
 Confidence Interval for 

Roughly 95% of the time, sample

means, X , will fall in this interval.

 - 2(St Error)   + 2(St Error)

7
 Confidence Interval for 
This is equivalent to saying that the population mean is within
two standard errors of the sample mean roughly 95% of the
time.
Based on this thinking, roughly 95% of the time, the population
mean falls in the interval:

X  2 S / n to X  2 S / n
We usually write this interval by enclosing the confidence limits
in parentheses and separating them with a comma:

8
 Confidence Interval for 
We say that an interval of this type contains the true mean roughly
with 95% confidence.
The use of +/-2 as the multiplier in the interval on the previous
slide gives a rather rough approximation to a 95% level
confidence interval.
In part, this is due to the fact that S, as an estimate of , varies
based on the sample, particularly when sample sizes are small.
To correctly compute a confidence interval, one must use “t-
values”.
These are values obtained from a statistical distribution known as
Student’s t distribution.

9
 n-1 p=0.600 0.750 0.900 0.950 0.975 0.990 0.995 0.999
1 0.325 Partial Student’s t Table
1.000 3.078 6.314 12.706 31.821 63.657 318.309
2 0.289 0.816 1.886 2.920 4.303 6.965 9.925 22.327
3 0.277 0.765 1.638 2.353 3.182 4.541 5.841 10.215
4 0.271 0.741 1.533 2.132 2.776 3.747 4.604 7.173
5 0.267 0.727 1.476 2.015 2.571 3.365 4.032 5.893
6 0.265 0.718 1.440 1.943 2.447 3.143 3.707 5.208
7 0.263 0.711 1.415 1.895 2.365 2.998 3.499 4.785
8 0.262 0.706 1.397 1.860 2.306 2.896 3.355 4.501
9 0.261 0.703 1.383 1.833 2.262 2.821 3.250 4.297
10 0.260 0.700 1.372 1.812 2.228 2.764 3.169 4.144
11 0.260 0.697 1.363 1.796 2.201 2.718 3.106 4.025
12 0.259 0.695 1.356 1.782 2.179 2.681 3.055 3.930
13 0.259 0.694 1.350 1.771 2.160 2.650 3.012 3.852
14 0.258 0.692 1.345 1.761 2.145 2.624 2.977 3.787
15 0.258 0.691 1.341 1.753 2.131 2.602 2.947 3.733
16 0.258 0.690 1.337 1.746 2.120 2.583 2.921 3.686
17 0.257 0.689 1.333 1.740 2.110 2.567 2.898 3.646
18 0.257 0.688 1.330 1.734 2.101 2.552 2.878 3.610
19 0.257 0.688 1.328 1.729 2.093 2.539 2.861 3.579
20 0.257 0.687 1.325 1.725 2.086 2.528 2.845 3.552
21 0.257 0.686 1.323 1.721 2.080 2.518 2.831 3.527
22 0.256 0.686 1.321 1.717 2.074 2.508 2.819 3.505
23 0.256 0.685 1.319 1.714 2.069 2.500 2.807 3.485
24 0.256 0.685 1.318 1.711 2.064 2.492 2.797 3.467
25 0.256 0.684 1.316 1.708 2.060 2.485 2.787 3.450
26 0.256 0.684 1.315 1.706 2.056 2.479 2.779 3.435
27 0.256 0.684 1.314 1.703 2.052 2.473 2.771 3.421
28 0.256 0.683 1.313 1.701 2.048 2.467 2.763 3.408
29 0.256 0.683 1.311 1.699 2.045 2.462 2.756 3.396
30 0.256 0.683 1.310 1.697 2.042 2.457 2.750 3.385
40 0.255 0.681 1.303 1.684 2.021 2.423 2.704 3.307
60 0.254 0.679 1.296 1.671 2.000 2.390 2.660 3.232
100 0.254 0.677 1.290 1.660 1.984 2.364 2.626 3.174
Critical values of the t-Distribution at n-1 degrees of freedom. 10
 n-1 p=0.600 0.750 0.900 0.950 0.975 0.990 0.995 0.999
1 0.325 1.000 3.078 6.314 12.706 31.821 63.657 318.309
2 0.289 0.816 1.886 2.920 4.303 6.965 9.925 22.327
3 0.277 0.765 1.638 2.353 3.182 4.541 5.841 10.215
4 0.271 Partial Student’s t Table
0.741 1.533 2.132 2.776 3.747 4.604 7.173
5 0.267 0.727 1.476 2.015 2.571 3.365 4.032 5.893
6 0.265 0.718 1.440 1.943 2.447 3.143 3.707 5.208
7 0.263 0.711 1.415 1.895 2.365 2.998 3.499 4.785
8 0.262 0.706 1.397 1.860 2.306 2.896 3.355 4.501
9 0.261 0.703 1.383 1.833 2.262 2.821 3.250 4.297
10 0.260 0.700 1.372 1.812 2.228 2.764 3.169 4.144
11 0.260 0.697 1.363 1.796 2.201 2.718 3.106 4.025
12 0.259 0.695 1.356 1.782 2.179 2.681 3.055 3.930
13 0.259 0.694 1.350 1.771 2.160 2.650 3.012 3.852
14 0.258 0.692 1.345 1.761 2.145 2.624 2.977 3.787
15 0.258 0.691 1.341 1.753 2.131 2.602 2.947 3.733
16 0.258 0.690 1.337 1.746 2.120 2.583 2.921 3.686
17 0.257 0.689 1.333 1.740 2.110 2.567 2.898 3.646
18 0.257 0.688 1.330 1.734 2.101 2.552 2.878 3.610
19 0.257 0.688 1.328 1.729 2.093 2.539 2.861 3.579
20 0.257 0.687 1.325 1.725 2.086 2.528 2.845 3.552
21 0.257 0.686 1.323 1.721 2.080 2.518 2.831 3.527
22 0.256 0.686 1.321 1.717 2.074 2.508 2.819 3.505
23 0.256 0.685 1.319 1.714 2.069 2.500 2.807 3.485
24 0.256 0.685 1.318 1.711 2.064 2.492 2.797 3.467
25 0.256 0.684 1.316 1.708 2.060 2.485 2.787 3.450
26 0.256 0.684 1.315 1.706 2.056 2.479 2.779 3.435
27 0.256 0.684 1.314 1.703 2.052 2.473 2.771 3.421
28 0.256 0.683 1.313 1.701 2.048 2.467 2.763 3.408
29 0.256 0.683 1.311 1.699 2.045 2.462 2.756 3.396
30 0.256 0.683 1.310 1.697 2.042 2.457 2.750 3.385
40 0.255 0.681 1.303 1.684 2.021 2.423 2.704 3.307
60 0.254 0.679 1.296 1.671 2.000 2.390 2.660 3.232
100 0.254 0.677 1.290 1.660 1.984 2.364 2.626 3.174
Critical values of the t-Distribution at n-1 degrees of freedom. 11
 Interpretation of Confidence Intervals
The graph shows 100 simulated 95% level confidence intervals for
samples of size 5 from a population with mean 3.
Intervals marked with bold endpoints did not capture  = 3.
Note that estimates of sigma in small samples are highly unstable.

12
 Confidence Interval for a Proportion
The idea behind a confidence interval for a population proportion
is similar to that for the mean of a continuous distribution.
Recall that the estimate of a population proportion is given by:

Number of successes in sample

p
Size of sample

Also, recall that the standard error for a sample proportion is

p (1  p)
n
where n is the sample size.

13
 Confidence Interval for a Proportion
The classical confidence interval for a proportion, based on the
Rule of Averages, is given by:

p(1
Here, z is a quintile of the standard normal distribution. The
value of z depends on the desired confidence level.

(p z
Typically used values of z are given in the
table to the right.
Confidence
Level
0.900
z
1.645

n
Note that these are consistent with the idea 0.950 1.960
0.980 2.326
that the sample proportion falls within two 0.990 2.576
standard errors of the true proportion about 0.995 2.807

95% of the time.

14