|
V24
Estimation Welcome to part seven of our
video series related to estimation. In this video, we are going to review the
Chi Square distribution and statistic as well as the
F distribution and statistic. I'm Renee Clark from the Swanson School of
Engineering at the University of Pittsburgh. Okay, so, first, the chi square
distribution. Okay, this distribution, as you'll recall, is used to
investigate or estimate sigma squared, which is our population variance. Okay,
the Chi Square distribution, of course, is always… is also used to estimate
Sigma, okay, which is the population standard deviation. So, either one… can
look at it either way, okay, and this is done by taking a sample from a
normal or normally distributed population. Okay, important
point about using the Chi Square distribution to make estimates is that your
population must be normally distributed. Okay, so, in other words, the
central limit theorem does not have a role in the Chi Square distribution. The
point estimate of Sigma squared is s squared, or the
sample variance. Okay, you'll also recall from earlier in our video series
that this particular statistic here, or the sample size minus one times the ratio of the sample variance to the population
variance, has a Chi Squared Distribution, we write Chi square like that, and
it has n minus 1 degrees of freedom. You recall that,
in general, the Chi square distribution is a continuous distribution. It's
skewed to the right, it's not symmetric, and the convention for using it,
including for the back of the book, is that alpha represents the area to the
right under the Chi Square curve… to the right of the critical value Chi
Square sub alpha. Okay, again, this should look
familiar to you. This is the critical values of the chi
Squared Distribution from the back of the book. Okay, here's a visual legend
that you should always reference when using a table. You'll see, again, alpha
is that area to the right… little bit difficult to say… to see, but critical
value sits along the x axis is Chi squared sub alpha degrees of freedom. Down
the left, values of alpha. Across the top, there are two pages of the…of
these critical values for the Chi Square distribution in the back of the book.
Values of Chi Square sit in the middle of the table, okay? Okay, now the F distribution. Okay,
the F distribution is used to investigate or estimate the ratio of the
variances from two independent populations, okay, which is given by the ratio
of Sigma 1^ 2 to Sigma 2^ 2. Okay, and, of course, the F distribution is also
used to investigate or estimate the ratio of the standard deviations from the
two independent populations. You can look at it either way. Okay, this is
done by taking samples from two independent, normally distributed populations.
Okay, again, an important point
about the F distribution, just as with the… the chi Square
distribution, your populations must be normally distributed in order to use
the F distribution. Okay, there is not a concept of the central limit theorem
with the F distribution. Okay, the point estimate for the ratio of the
population variances is going to be, of course, the ratio of the two sample
variances, S1^2, S2^2. Okay, you'll recall that this
statistic right here is distributed according to the F distribution. Okay, so,
that statistic is in the numerator… has a numerator of… for the first
population, the ratio of the sample variance to the population variance. Okay,
and then, in the denominator, the ratio from the… for the second population,
the ratio of the sample variance to the population variance. Okay, and so,
this particular statistic, here, can also be
rewritten in this way just by rearranging some terms. Okay, the… the… there
are two sets of…there are two degrees of freedom associated with the F
distribution: from the first population, N1 minus one, and from the second
population, N2 minus 1. So, each sample size minus one. Recall that the F
distribution is also a continuous distribution like the chi squared, also
skewed right, not symmetric. Okay, the convention in…including… for the table
in the back of the book is alpha represents area to the right… under the F
curve… to the right of the critical value for f that sits along the x-axis. We
call that F sub Alpha, so that Alpha is area to the
right. Okay, this should look familiar to you. This is the table of critical
values of the F distribution that's in the back of the book. There are four…
four total pages of these critical values in the back of the book. Across the
top is your nu one, or N1
-1 (first degree of Freedom) Second degree of Freedom, nu 2 sits down the left hand side. That is your N2 minus one. Okay, critical values for f sit
in the middle of the table. These are your F… it's… sub
Alpha. This particular table is for Alpha of .05
because of that subscript, there. Okay, and, again, always take note of your
visual legend when you're using a table. F sub Alpha
sits right there, do that in blue. F sub Alpha- a
little difficult to see there, but F sub Alpha is your critical value. It sits
along the x-axis. Alpha, of course, by convention, you… in looking… in the
visual legend represents your area to the right under the curve… curve to the
right of your F sub Alpha. Okay, we also have tables
in the back of the book for an alpha of .01 as well.
We wish to thank the National
Science Foundation under Grant 233582 for supporting our work. Thank you for watching. |