Understanding the chi-squared distribution

1. Formula
   Observation

   Slide shows $\chi^2 = Z_1^2 + Z_2^2 + \cdots + Z_r^2$ and later $\chi^2(r)$.

2. Audio
   Observation

   Speaker says the sum is the random variable chi-squared and that its sampling distribution is called the chi-squared distribution with r degrees of freedom.

$\chi^2$

Random variable equal to the sum of squared z-scores; also used as the name of its sampling distribution.

Nonnegative values for the distribution support; the slide states the probability density function is zero for $\chi^2 \le 0$.

1. Formula
   Observation

   Slide writes $Z_1, Z_2, \ldots, Z_r$ and uses $Z_i^2$ in the sum.

2. Audio
   Observation

   Speaker says these are the z-scores of the observations.

$Z_i$

The z-score of the i-th independent numerical observation.

Continuous random variable according to the slide.

1. Formula
   Observation

   Slide uses $Z_r$ as the last term and $\chi^2(r)$ for the distribution.

2. Audio
   Observation

   Speaker says r is the number of z-scores going in, the number of independent observations.

r

Number of independent observations or z-scores being summed; degrees of freedom of the chi-squared distribution.

Positive integer count of terms in the displayed finite sum.

1. Formula
   Observation

   Slide states: The probability density function for $\chi^2$ is zero for $\chi^2 \le 0$ and is skewed to the right for any $r$.

2. Audio
   Observation

   Speaker says it can be described with a probability density function and that it is only positive for chi-squared values at least zero.

1. The explicit algebraic formula for the density is not shown in this clip.

$\text{probability density function for }\chi^2$

Density describing the continuous sampling distribution of the chi-squared random variable.

Zero for $\chi^2 \le 0$; positive/right-skewed behavior stated for $\chi^2>0$ and any $r$.

1. Caption evidence
   Observation

   Slide text uses $Z_i$ in “Since each $Z_i$ is a continuous random variable…”

2. Audio
   Observation

   Speaker refers to “the z scores” and “standard normal distributions for those z scores.”

$Z_i$

Individual standard normal random variables (z-scores) whose squares are summed to form $\chi^2$.

Continuous real-valued random variable.

1. Caption evidence
   Observation

   Slide text defines $\chi^2 = Z_1^2 + \cdots + Z_r^2$ and discusses its probability density function.

2. Audio
   Observation

   Speaker repeatedly says “chi-squared distribution.”

$\chi^2$

Chi-squared random variable, defined as the sum of squares of independent standard normal random variables.

Nonnegative real values; strictly positive support as stated by the speaker.

1. Caption evidence
   Observation

   Slide text uses $r$ in $\chi^2 = Z_1^2 + \cdots + Z_r^2$, $\chi^2(r)$, and Fact 2 condition $r \geq 2$.

2. Audio
   Observation

   Speaker calls $r$ the number of “degrees of freedom.”

r

Degrees of freedom parameter of the chi-squared distribution; also the number of squared standard normal terms in the definition.

Positive integer or real-valued shape parameter depending on convention; video treats it as count of degrees of freedom.

1. Caption evidence
   Observation

   Slide text: “When drawing samples of size $n$ from a normal distribution…”

2. Audio
   Observation

   Speaker says “samples of size n.”

n

Sample size used in the variance-testing application.

Positive integer.

1. Caption evidence
   Observation

   Slide text: “hypothesized variance $\sigma^2$” and formula $(n - 1)S^2/\sigma^2$.

2. Audio
   Observation

   Speaker says “hypothesized variance sigma squared.”

$\sigma^2$

Hypothesized population variance in the normal model for significance testing of variance.

Positive real number.

1. Caption evidence
   Observation

   Slide text: “where $S^2$ is the sample variance.”

2. Audio
   Observation

   No explicit spoken definition beyond the slide context.

$S^2$

Sample variance.

Nonnegative real number.

1. Formula
   Observation

   The slide text repeatedly uses $\chi^2$ and $\chi^2(r)$ to denote the chi-squared distribution and a chi-squared distribution with $r$ degrees of freedom.

2. Audio
   Observation

   The speaker says "chi-squared distribution" and "chi-squared distribution with five degrees of freedom."

$\chi^2$

Chi-squared distribution; when written as $\chi^2(r)$ it denotes the chi-squared distribution with $r$ degrees of freedom.

Distribution family used in inferential statistics and probability calculations.

1. Formula
   Observation

   Slide text: "When drawing samples of size $n$ from a normal distribution..."

2. Audio
   Observation

   The speaker reads the formula involving $(n-1)S^2/\sigma^2$ and refers to $n-1$ in the degrees of freedom.

n

Sample size in the variance-testing application.

Positive integer sample size in a random sample from a normal distribution.

1. Formula
   Observation

   Slide: If the z-scores of the observations are $Z_1, Z_2, \ldots, Z_r$, then we define $\chi^2 = Z_1^2 + Z_2^2 + \cdots + Z_r^2$.

2. Audio
   Observation

   Speaker explains taking each observation, computing a z-score by subtracting expected value and dividing by standard deviation, squaring each z-score, and adding them all up.

The clip defines $\chi^2$ as the total squared deviation of independent numerical observations from their expected values, measured through z-scores. Each observation is standardized into a z-score, then squared, and the squares are summed over $r$ independent observations.

1. The observations are independent numerical observations.

2. Each $Z_i$ is the z-score of an observation.

3. The sum contains $r$ squared z-scores.

1. Computing a z-score for each observation

1. Audio
   Observation

   Speaker says: compute a z-score, so you subtract the expected value and divide by the standard deviation.

2. Formula
   Observation

   Slide names the resulting quantities $Z_1, Z_2, \ldots, Z_r$ as the z-scores of the observations.

1. The explicit formula $Z=(X-E[X])/\sigma$ is spoken but not written on the slide.

Before forming $\chi^2$, each observation is converted to a z-score by subtracting its expected value and dividing by its standard deviation. This measures how far one individual observation lies from its expected value in standard-deviation units.

1. Each observation has an expected value.

2. Each observation has a standard deviation used for scaling.

1. Formula
   Observation

   Slide: The sampling distribution of the random variable $\chi^2$ is called the $\chi^2$-distribution with $r$ degrees of freedom, or $\chi^2(r)$ for short.

2. Audio
   Observation

   Speaker says the sampling distribution of this random variable chi-squared is called the chi-squared distribution with r degrees of freedom.

The sampling distribution of the random variable $\chi^2$ is named the chi-squared distribution with $r$ degrees of freedom, abbreviated $\chi^2(r)$. In this clip, $r$ is identified as the number of independent z-scores included in the sum.

1. There are $r$ independent z-scores in the defining sum.

1. Definition of the chi-squared random variable
2. Meaning of r as degrees of freedom in this construction

1. Audio
   Observation

   Speaker says: So here is r the number of z-scores that are going in, the number of those independent observations.

2. Formula
   Observation

   Slide uses $r$ as the upper index in $Z_r$ and in $\chi^2(r)$.

1. The clip does not discuss more general degrees-of-freedom adjustments beyond counting the independent z-scores.

In the presented construction, the degrees of freedom parameter $r$ is the number of independent observations, equivalently the number of z-scores being squared and summed.

1. The $Z_i$ terms correspond to independent observations.

1. Definition of the chi-squared random variable

1. Formula
   Observation

   Slide: Since each $Z_i$ is a continuous random variable, so is $\chi^2 = Z_1^2 + \cdots + Z_r^2$.

2. Audio
   Observation

   Speaker says since each z-score is a continuous random variable, so is the sum of the squares of them.

Because each $Z_i$ is continuous, the sum of their squares is also continuous. This justifies describing the sampling distribution of $\chi^2$ using a probability density function.

1. Each $Z_i$ is a continuous random variable.

1. Definition of the chi-squared random variable

1. Formula
   Observation

   Slide: The probability density function for $\chi^2$ is zero for $\chi^2 \le 0$ and is skewed to the right for any $r$.

2. Audio
   Observation

   Speaker says the probability density function is only going to be positive for chi-squared values that are at least zero.

1. The exact density expression is absent from this clip.

The clip states that the probability density function of $\chi^2$ is zero on the nonpositive side and is right-skewed for any $r$. This matches the fact that $\chi^2$ is a sum of squared quantities and therefore cannot be negative.

1. $x$ denotes a possible value of the chi-squared random variable.

2. The statement applies for any $r$ shown in the slide.

1. Definition of the chi-squared random variable
2. Chi-squared is a continuous random variable

1. Caption evidence
   Observation

   Slide states: “Since each $Z_i$ is a continuous random variable, so is $\chi^2 = Z_1^2 + \cdots + Z_r^2$.”

2. Audio
   Observation

   Speaker explains that extreme values in each standard normal distribution become decreasingly likely, motivating the shape of the resulting distribution.

The video defines $\chi^2$ as the sum of the squares of $r$ standard normal random variables $Z_1,\dots,Z_r$. Because each $Z_i$ is continuous, the resulting $\chi^2$ is also continuous.

1. Each $Z_i$ is a standard normal random variable.

2. There are $r$ such variables being squared and summed.

1. $Z_i$
2. $\chi^2$
3. r

1. Caption evidence
   Observation

   Slide states: “The probability density function for $\chi^2$ is zero for $\chi^2 \leq 0$ and is skewed to the right for any $r$.”

2. Audio
   Observation

   Speaker says the distribution is “skewed to the right” because very extreme values in the standard normal components are decreasingly likely.

3. Diagram
   Observation

   Graph labeled $\chi^2(5)$ shows a curve starting at 0, rising to a peak near 3, then tapering with a long right tail.

The probability density function of $\chi^2$ is zero on $(-\infty,0]$ and is right-skewed for every $r$. The speaker justifies the skew intuitively by noting that extreme values in the underlying standard normal variables become increasingly unlikely.

1. Applies for any degrees of freedom $r$.

2. Support is nonnegative; speaker describes possible outcomes as strictly positive.

1. Definition of the chi-squared random variable
2. $\chi^2$
3. r

1. Caption evidence
   Observation

   Slide text: “Fact 1. The expected value of $\chi^2(r)$ is $r$.”

2. Audio
   Observation

   Speaker says: “The expected value of the chi-squared distribution with r degrees of freedom is just r,” then motivates this from the definition as a sum of squared z-scores with standard deviation 1.

For a chi-squared distribution with $r$ degrees of freedom, the mean equals $r$. The speaker gives an intuitive derivation: each squared standard normal contributes an average amount equal to 1, so $r$ such contributions give expected total $r$.

1. $\chi^2(r)$ denotes a chi-squared distribution with $r$ degrees of freedom.

1. Definition of the chi-squared random variable
2. $\chi^2$
3. r

1. Caption evidence
   Observation

   Slide text: “Fact 2. The mode (peak) of $\chi^2(r)$ is $r - 2$ if $r \geq 2$ and zero otherwise.”

2. Audio
   Observation

   Speaker says the peak is at $r-2$ as long as $r$ is at least 2, and otherwise it is zero.

The location of the peak of the chi-squared density depends on $r$: if $r\ge 2$, the mode is $r-2$; otherwise the mode is 0.

1. Applies to the density of $\chi^2(r)$.

1. $\chi^2$
2. r

1. Caption evidence
   Observation

   Slide text: “When $r$ is large, the distribution $\chi^2(r)$ is approximately normal. This is already visible when $r = 50$, as pictured below.” and “If you look closely you can still see the skew in this plot, however.”

2. Audio
   Observation

   Speaker explicitly invokes the central limit theorem: “By the central limit theorem, when the number of degrees of freedom gets large, the distribution chi-squared of r is going to become approximately normal.”

3. Diagram
   Observation

   Graph labeled $\chi^2(50)$ shows a bell-shaped curve centered near 50 with slight residual right skew.

As the degrees of freedom $r$ grow, the chi-squared distribution becomes approximately normal by the central limit theorem. The video illustrates this with $r=50$, where the curve looks bell-shaped but still retains a small rightward skew.

1. $r$ is large.

2. Approximation is qualitative in the video; no standardized normal parameters are written on screen.

1. Definition of the chi-squared random variable
2. r

1. Caption evidence
   Observation

   Slide text: “Significance testing for variance. When drawing samples of size $n$ from a normal distribution with hypothesized variance $\sigma^2$, the sampling distribution of $(n - 1)S^2/\sigma^2$ is $\chi^2(n - 1)$, where $S^2$ is the sample variance.”

2. Audio
   Observation

   Speaker introduces this as one of the most common uses of the chi-squared distribution in inferential statistics.

In inference about variance under a normal model, the statistic $(n-1)S^2/\sigma^2$ has a chi-squared distribution with $n-1$ degrees of freedom. This is presented as a major application of the chi-squared family.

1. Samples are drawn from a normal distribution.

2. $\sigma^2$ is the hypothesized population variance.

3. $S^2$ is the sample variance.

4. $n$ is the sample size.

1. Definition of the chi-squared random variable
2. n
3. $\sigma^2$
4. $S^2$

1. Formula
   Observation

   Slide: For instance, when all of the outcomes of the independent variables are exactly at their mean, $\chi^2$ is exactly zero.

2. Audio
   Observation

   Speaker says if all observations were exactly at the expected value, then the chi-squared statistic would be zero.

If every independent observation is exactly at its expected value, then $\chi^2=0$.

1. The observations are independent.

2. Each $Z_i$ is the z-score of an observation.

3. Every observation equals its expected value.

For all observations in the collection.

1. Formula
   Observation

   Slide: As the individual results get more extreme, $\chi^2$ gets larger.

2. Audio
   Observation

   Speaker says if the results were further away, more extreme, then that chi-squared value would get bigger and bigger.

1. The slide does not formalize a monotonicity theorem for arbitrary changes; it gives an intuitive directional statement.

As individual results become more extreme relative to their expected values, the value of $\chi^2$ increases.

1. $\chi^2$ is formed from squared z-scores of the observations.

Qualitative statement about individual results becoming more extreme.

1. Formula
   Observation

   Slide: By squaring the z-scores before summing, we insure that low results don't cancel out high ones.

2. Audio
   Observation

   Speaker says the idea behind squaring all of these z-scores is to ensure that low results do not somehow cancel out or balance out with the high results.

Squaring the z-scores before summing prevents low and high results from cancelling each other out in the total measure of deviation.

1. Deviations are represented by signed z-scores before squaring.

General design rationale for the definition of $\chi^2$.

1. Audio
   Observation

   Speaker says a z-score is supposed to measure how far an individual observation is from an expected value, either in the positive or negative direction.

1. No explicit formula for signed z-scores is shown on screen.

A z-score measures how far an individual observation is from its expected value in either the positive or negative direction.

1. The observation has an expected value and standard deviation.

For an individual observation.

1. Audio
   Observation

   Speaker says notice the name of the random variable is the same as the name of the distribution; context usually makes the difference clear.

The name $\chi^2$ is used both for the random variable defined by the sum of squared z-scores and for its sampling distribution.

1. The discussion distinguishes the random variable from its sampling distribution by context.

Naming convention within this lecture.

1. Audio
   Observation

   Speaker says: “you can see that its support, the set of possible outcomes, is strictly positive.”

2. Caption evidence
   Observation

   Slide text says the pdf is zero for $\chi^2 \leq 0$.

1. The slide wording allows value 0 in the support boundary, while the speaker verbally says “strictly positive”; this is a phrasing difference rather than a mathematical contradiction in the displayed graph.

The chi-squared distribution has support on nonnegative values, with density zero for $\chi^2 \leq 0$; the speaker describes the possible outcomes as strictly positive.

1. $\chi^2$ is defined as a sum of squared standard normal variables.

For all values in the domain of the density.

1. Caption evidence
   Observation

   Slide text: “…is skewed to the right for any $r$.”

2. Audio
   Observation

   Speaker says the distribution is skewed to the right and points to the $\chi^2(5)$ graph as typical.

For any $r$, the chi-squared density is right-skewed.

1. $r$ is the degrees-of-freedom parameter of $\chi^2(r)$.

For all $r$ discussed in the video.

1. Audio
   Observation

   Speaker says: “By the central limit theorem, when the number of degrees of freedom gets large, the distribution chi-squared of r is going to become approximately normal.”

2. Caption evidence
   Observation

   Slide text: “When $r$ is large, the distribution $\chi^2(r)$ is approximately normal.”

When $r$ is large, $\chi^2(r)$ is approximately normal.

1. $\chi^2(r)$ is a sum of $r$ squared standard normal variables.

2. $r$ is large.

Asymptotic statement in $r$; the video presents it qualitatively rather than with explicit limiting notation.

1. Formula
   Observation

   Slide text states that the sampling distribution of $(n-1)S^2/\sigma^2$ is $\chi^2(n-1)$ under the listed conditions.

2. Audio
   Observation

   The speaker reads the claim verbatim.

When drawing samples of size $n$ from a normal distribution with hypothesized variance $\sigma^2$, the sampling distribution of $(n-1)S^2/\sigma^2$ is $\chi^2(n-1)$, where $S^2$ is the sample variance.

1. Samples have size $n$.

2. The parent distribution is normal.

3. $\sigma^2$ is the hypothesized variance.

4. $S^2$ is the sample variance.

For samples of size $n$ from a normal distribution with hypothesized variance $\sigma^2$.

1. Formula
   Observation

   Slide text states that the sampling distribution of $\sum (O-E)^2/E$ is approximately $\chi^2(n-1)$ under the listed conditions.

2. Audio
   Observation

   The speaker reads the approximation claim aloud.

1. The slide does not state additional regularity conditions for the approximation beyond those shown.

When a categorical variable is hypothesized to have a certain distribution, the sampling distribution of $\sum \frac{(O-E)^2}{E}$ is approximately $\chi^2(n-1)$, where $n$ is the number of categories, $O$ is the observed count in each category, and $E$ is the expected count under the hypothesized distribution.

1. The variable is categorical.

2. A specific distribution is hypothesized for the categories.

3. $n$ is the number of categories.

4. $O$ is the observed count in each category.

5. $E$ is the expected count under the hypothesized distribution.

For a categorical variable hypothesized to have a certain distribution.

1. Formula
   Observation

   Slide text: "For any given value $x$, the cdf $F(x)$ gives the probability of randomly getting a value less than or equal to $x$ in the appropriate $\chi^2$ distribution. Technically, $F(x)=P(X\le x)$ in $\chi^2(r)$".

2. Audio
   Observation

   The speaker states the same interpretation in words.

For any given value $x$, the cumulative distribution function $F(x)$ gives the probability of randomly getting a value less than or equal to $x$ in the appropriate $\chi^2$ distribution; technically, $F(x)=P(X\le x)$ in $\chi^2(r)$.

1. The distribution is a chi-squared distribution.

2. $x$ is a given value of interest.

3. $r$ is the degrees-of-freedom parameter.

For any given value $x$.

1. Formula
   Observation

   Slide introduces $Z_1, Z_2, \ldots, Z_r$ and then displays $\chi^2 = Z_1^2 + Z_2^2 + \cdots + Z_r^2$.

2. Audio
   Observation

   Speaker describes computing z-scores, squaring them, and adding them all up to obtain the random variable chi-squared.

1. Expression

   $$Z_i = \frac{X_i - E[X_i]}{\sigma_{X_i}}$$
   Explanation

   Each observation is standardized into a z-score by subtracting its expected value and dividing by its standard deviation.

   Justification

   Spoken definition of z-score in the clip.

   Shown in the video
2. Expression

   $$Z_1^2, Z_2^2, \ldots, Z_r^2$$
   Explanation

   Each z-score is squared so that deviations contribute nonnegative amounts.

   Justification

   Slide and audio state that the z-scores are squared before summing.

   Shown in the video
3. Expression

   $$\chi^2 = Z_1^2 + Z_2^2 + \cdots + Z_r^2$$
   Explanation

   The squared z-scores are added to form the total deviation measure $\chi^2$.

   Justification

   Displayed definition on the slide.

   Shown in the video

The chi-squared random variable is the sum of the squared z-scores of $r$ independent observations.

1. Formula
   Observation

   Slide states that when all outcomes are exactly at their mean, $\chi^2$ is exactly zero.

2. Audio
   Observation

   Speaker repeats that if all observations were exactly at the expected value, the chi-squared statistic would be zero.

1. Expression

   $$X_i = E[X_i]$$
   Explanation

   Assume every observation is exactly at its expected value.

   Justification

   Stated example condition on the slide.

   Shown in the video
2. Expression

   $$Z_i = 0$$
   Explanation

   Then each z-score is zero because the numerator of the standardization is zero.

   Justification

   Derived from the spoken z-score construction.

   Derived from the video
3. Expression

   $$\chi^2 = 0^2 + 0^2 + \cdots + 0^2 = 0$$
   Explanation

   Summing the squared zero z-scores gives zero.

   Justification

   Direct substitution into the displayed definition of $\chi^2$.

   Derived from the video

When all observations are exactly at their means, $\chi^2=0$.

1. Formula
   Observation

   Slide: Since each $Z_i$ is a continuous random variable, so is $\chi^2 = Z_1^2 + \cdots + Z_r^2$.

2. Formula
   Observation

   Next slide line: The probability density function for $\chi^2$ is zero for $\chi^2 \le 0$ and is skewed to the right for any $r$.

3. Audio
   Observation

   Speaker says because each z-score is continuous, so is the sum of squares, and we can describe it with a probability density function.

1. The clip asserts continuity and density description but does not prove closure properties in detail.

1. Expression

   $$Z_i \text{ is continuous}$$
   Explanation

   Each z-score is identified as a continuous random variable.

   Justification

   Explicit slide statement.

   Shown in the video
2. Expression

   $$\chi^2 = Z_1^2 + \cdots + Z_r^2 \text{ is continuous}$$
   Explanation

   The sum of squared continuous variables is described as continuous.

   Justification

   Explicit slide statement and speaker explanation.

   Shown in the video
3. Expression

   $$f_{\chi^2}(x)=0 \text{ for } x \le 0$$
   Explanation

   Because $\chi^2$ is continuous, the clip says it can be described by a probability density function, whose support begins at zero.

   Justification

   Slide statement plus spoken explanation.

   Shown in the video

The clip uses continuity of the $Z_i$ to justify describing the sampling distribution of $\chi^2$ with a probability density function supported on nonnegative values.

1. Audio
   Observation

   Speaker motivates Fact 1 from the definition: chi-squared is the sum of squares of individual z-scores, each standard normal variable has standard deviation 1, and with $r$ observations the expected difference totals $r$.

2. Caption evidence
   Observation

   Slide states Fact 1: expected value of $\chi^2(r)$ is $r$.

1. The spoken argument is intuitive rather than a fully formal proof; it does not explicitly write out $E[Z_i^2]=\operatorname{Var}(Z_i)+(E[Z_i])^2=1$ on screen.

1. Expression

   $$$$
   Explanation

   Start from the definition $\chi^2 = Z_1^2+\cdots+Z_r^2$.

   Justification

   Given by the slide definition of the chi-squared random variable.

   Shown in the video
2. Expression

   $$$$
   Explanation

   Each $Z_i$ is standard normal, so its standard deviation is 1.

   Justification

   Stated by the speaker in the audio.

   Shown in the video
3. Expression

   $$$$
   Explanation

   The speaker interprets each squared z-score as contributing an average amount equal to 1 to the total.

   Justification

   Intuitive explanation given verbally; not written as a formal algebraic step on screen.

   Shown in the video
4. Expression

   $$E[\chi^2(r)] = r$$
   Explanation

   Adding $r$ such contributions yields expected value $r$.

   Justification

   Conclusion stated as Fact 1 on the slide and repeated in the narration.

   Shown in the video

The expected value of a chi-squared distribution with $r$ degrees of freedom is $r$.

1. Audio
   Observation

   Speaker explicitly cites the central limit theorem to explain why large-$r$ chi-squared looks approximately normal.

2. Diagram
   Observation

   The $\chi^2(50)$ plot visually resembles a bell curve more closely than the $\chi^2(5)$ plot.

1. The video does not show the standardized form $(\chi^2(r)-r)/\sqrt{2r}$ or any explicit convergence statement.

1. Expression

   $$$$
   Explanation

   $\chi^2(r)$ is a sum of $r$ squared standard normal variables.

   Justification

   Definition shown earlier on the slide.

   Shown in the video
2. Expression

   $$$$
   Explanation

   As $r$ grows, sums of many random components tend toward normality by the central limit theorem.

   Justification

   Explicitly invoked by the speaker.

   Shown in the video
3. Expression

   $$$$
   Explanation

   The plotted case $r=50$ already appears close to bell-shaped.

   Justification

   Visible in the graph and noted in the slide text.

   Shown in the video
4. Expression

   $$$$
   Explanation

   A slight right skew remains visible if inspected closely.

   Justification

   Stated in the slide caption and by the speaker.

   Shown in the video

For large $r$, $\chi^2(r)$ is approximately normal, though residual right skew can remain visible.

1. Formula
   Observation

   Slide: For instance, when all of the outcomes of the independent variables are exactly at their mean, $\chi^2$ is exactly zero.

2. Audio
   Observation

   Speaker gives the same example verbally.

What is $\chi^2$ when every independent observation is exactly at its expected value?

1. Observations are independent.

2. Each observation is exactly at its mean/expected value.

3. $\chi^2$ is the sum of squared z-scores.

Determine the value of $\chi^2$ in this case.

1. Expression

   $$Z_i = 0 \text{ for each } i$$
   Explanation

   If an observation equals its expected value, its z-score is zero.

   Justification

   Follows from the spoken definition of z-score.

   Derived from the video
2. Expression

   $$\chi^2 = 0^2 + 0^2 + \cdots + 0^2$$
   Explanation

   Substitute all zero z-scores into the definition of $\chi^2$.

   Justification

   Uses the displayed formula for $\chi^2$.

   Derived from the video
3. Expression

   $$\chi^2 = 0$$
   Explanation

   The sum of zeros is zero.

   Justification

   Arithmetic simplification.

   Derived from the video

$\chi^2=0$.

Matches the slide statement that in this situation $\chi^2$ is exactly zero.

1. Formula
   Observation

   Slide: As the individual results get more extreme, $\chi^2$ gets larger.

2. Audio
   Observation

   Speaker says if results were further away, more extreme, then the chi-squared value would get bigger and bigger.

1. No numerical data set is provided; this is a qualitative example.

How does $\chi^2$ behave when individual results move farther from their expected values?

1. $\chi^2$ is a sum of squared z-scores.

2. Individual results become more extreme relative to expected values.

Describe the effect on $\chi^2$.

1. Expression

   $$|Z_i| \text{ becomes larger}$$
   Explanation

   More extreme observations correspond to larger absolute z-scores.

   Justification

   Interpretation of the spoken description of distance from expected value.

   Derived from the video
2. Expression

   $$Z_i^2 \text{ becomes larger}$$
   Explanation

   Squaring a larger absolute deviation yields a larger nonnegative contribution.

   Justification

   Algebraic property of squaring real numbers.

   Derived from the video
3. Expression

   $$\chi^2 \text{ gets larger}$$
   Explanation

   Since $\chi^2$ sums these squared contributions, the total increases.

   Justification

   Slide statement and definition of $\chi^2$.

   Shown in the video

$\chi^2$ gets larger as individual results become more extreme.

Consistent with the slide sentence and the speaker's verbal explanation.

1. Diagram
   Observation

   Graph labeled $\chi^2(5)$ on axes with x-values 0 to 12 and y-values up to 0.2.

2. Audio
   Observation

   Speaker says: “The graph of chi-squared of five is typical.”

Illustrate the typical shape of a chi-squared distribution using $\chi^2(5)$.

1. Distribution is $\chi^2(5)$.

2. Graph shows density versus outcome value.

Identify qualitative features: support, peak location, and skew.

1. Expression

   $$$$
   Explanation

   Read the horizontal axis as possible values of the random variable and the vertical axis as density.

   Justification

   Standard graph interpretation supported by the displayed axes labels.

   Shown in the video
2. Expression

   $$$$
   Explanation

   The curve starts at 0 and remains nonnegative thereafter.

   Justification

   Matches the slide statement that the pdf is zero for $\chi^2\le 0$.

   Shown in the video
3. Expression

   $$$$
   Explanation

   The peak occurs near 3, consistent with the later mode rule $r-2=5-2=3$.

   Justification

   Visual reading of the graph plus Fact 2 stated later in the clip.

   Derived from the video
4. Expression

   $$$$
   Explanation

   The right tail extends farther than the left side, showing right skew.

   Justification

   Visible in the plotted curve and described verbally by the speaker.

   Shown in the video

The $\chi^2(5)$ graph is a right-skewed density on nonnegative values with peak near 3.

Consistent with the slide text on support/skew and with Fact 2 giving mode $r-2$ for $r\ge2$.

1. Diagram
   Observation

   Graph labeled $\chi^2(50)$ on axes with x-values roughly 25 to 75 and y-values up to 0.06.

2. Caption evidence
   Observation

   Slide text: “This is already visible when $r = 50$, as pictured below.” and “If you look closely you can still see the skew in this plot, however.”

3. Audio
   Observation

   Speaker says the sketch shows $r=50$ and that squinting reveals a bit of rightward skew.

Show how the chi-squared distribution changes when $r$ is large using $\chi^2(50)$.