Allow each subject to have a separate baseline and slope through a correlated random-intercept and random-slope model:
The subject random effects are independent of all measurement errors, and subjects are independent. The within-person covariance between times is , with an additional when the same reading is used twice.
Prefer the model with a random slope. The maximum likelihood estimation refits improve twice the log-likelihood by while adding two covariance statistical parameters. Both the Akaike information criterion ( to ) and Bayesian information criterion ( to ) strongly favour it. The printed nominal likelihood-ratio test is overwhelming. The usual reference chi-squared distribution with two statistical degrees of freedom is not an exact regular calibration, since zero slope variance is a boundary and the intercept–slope correlation is then unidentified; a design-specific parametric bootstrap could calibrate it. This qualification does not undermine the substantial descriptive improvement shown by both information criteria.
The preferred fit estimates population mean baseline strength kg and weekly gain kg/week. The between-person baseline standard deviation is kg, and the between-person slope standard deviation is kg/week. Their estimated correlation is weakly positive, giving random-effect covariance about kg/week; its uncertainty is not supplied. The measurement-error standard deviation is kg. The printed instead describes the correlation between the estimated fixed effects, not between the subject random effects. These statistical parameter estimates come from restricted maximum likelihood; the model comparison uses the separate maximum likelihood estimation fits.
The population mean fitted trajectory is kg. Hence
Using the reported slope standard error, an approximate 95% confidence interval for the population mean gain is kg; a Student t confidence interval with 499 statistical degrees of freedom is almost identical.
Successful individuals can improve far more than the average. Their latent ten-week gains have fitted normal distribution
A clear estimate for an upper-performing group uses a between-person slope quantile: the 95th percentile is kg, and the 97.5th percentile is about kg. The upper 5% of fitted underlying gains begin around 73 kg. These are person-to-person performance quantiles, not confidence intervals for the population mean. Identifying the best observed trainee would require that person's data or fitted subject random effects; the aggregate output cannot identify a literal maximum. If performance means the observed difference of endpoint readings, add the measurement-error variance to the latent gain variance.
For subject and , the random-intercept linear mixed model is
All subject random effects and measurement errors are mutually independent. The fixed effects describe the population mean baseline and weekly gain. Conditional on , a person's observations have independent normal distributions; after integrating out , their covariance is at distinct weeks and their common variance is . Therefore their correlation is .
This Gaussian linear mixed model permits correlated repeated readings while keeping different subjects independent. The fitted standard deviations are kg and kg, giving within-person correlation about . All persons still have the same latent weekly slope in this model.
Measurements from the same person share baseline strength and likely share an individual response to training. An ordinary linear model with independent errors treats all 600 readings as independent conditional on week; that fails to represent their within-person correlation. The regression residuals can remain associated even when the population mean is linear in time. A random intercept captures differences in baseline strength, and a random slope can additionally describe differences in progress.
Under the more plausible Quasi-Poisson regression, the year Wald statistic is , with reported p-value . An approximate 95% confidence interval is
Thus the estimated fall is about 9.35%, but the data do not establish a reduction at the 5% level after allowing for overdispersion. The interval allows both a sizeable reduction and a small increase. The before–after comparison also lacks a contemporaneous randomized control, so causal inference about the campaign would require addressing other changes between years. The small Poisson regression p-value is not sufficient evidence of a campaign effect when its variance assumption is unsuitable.
The Quasi-Poisson regression retains the same conditional mean but permits
This is a mean–variance function specification through quasi-likelihood; it does not assign a full probability distribution to each count. For independent observations, the quasi-score equation is proportional to , so the mean statistical parameter estimates equal those from Poisson regression. The output estimates through the Pearson dispersion estimator, and inflates the standard errors by approximately .
The large residual deviance relative to 728 residual statistical degrees of freedom also signals substantial overdispersion. Daily weather, traffic and other omitted conditions may produce greater count variation than a homogeneous Poisson distribution allows. The Quasi-Poisson regression accounts for that extra marginal variance. It still requires a correct conditional mean and an appropriate independence assumption; a common dispersion parameter alone does not repair serial correlation.
The coefficient labelled yr2 identifies year as a factor. With in the first year and in the second, the Poisson regression assumes independent daily counts conditional on year,
The unknown mean statistical parameters are the first-year log daily rate and the second-year log rate ratio ; the dispersion parameter is fixed at one. Approximate 95% Wald confidence intervals are
Exponentiating yields the first-year fitted daily mean , with confidence interval , and
The second-year fitted daily mean is . Thus this Poisson regression estimates a 9.35% fall, with an approximate interval for the percentage fall from 3.66% to 14.70%. These confidence intervals rely on the Poisson distribution and independence assumptions.
For and , the Poisson distribution has mass
Match this to the exponential dispersion family with
Then and . The canonical link function expresses the natural parameter in terms of the mean, so the Poisson canonical link is . Its conditional variance equals its conditional mean.
An exponential dispersion family has density or mass function, relative to a fixed measure,
Here is the natural parameter, is the cumulant function, and is the dispersion parameter. Assume the natural parameter lies in the interior of its domain and derivatives can pass through the normalizing integral. Differentiating normalization once and twice gives
The variance function is , where the mean-to-natural-parameter inverse exists. The dispersion parameter may be fixed, as in a Poisson distribution, rather than estimated.
The analysis of variance provides evidence of a chocolate effect both with day included (, p-value ) and with day omitted (, p-value ). Because every chocolate–day cell has the same replication, balanced factorial orthogonality separates the two main effects; the chocolate row is meaningful despite being entered first.
The chocolate-only fitted group means are approximately boards for A, B, C, D respectively. Thus A has the largest fitted lecturing speed and C the smallest. Relative to A, the fitted differences are boards. The printed individual Student t-tests give strong evidence for the A–C contrast (p-value ); B and D versus A have p-values and , respectively. Those latter contrasts are not significant at 5%, and the output does not test all other pairwise comparisons. Simultaneous claims would require accounting for multiple hypothesis testing.
There is little evidence of a day effect after adjusting for chocolate. A chocolate-only normal linear model is therefore a reasonable simpler summary, with residual standard deviation about boards and explained variation . This is an association under the additive statistical model; a causal claim would additionally require an appropriate assignment of chocolate and checks of the regression residuals and possible interaction terms.
Test for both nonreference days, against at least one nonzero day fixed effect, conditional on chocolate. There are two added regression coefficients and residual statistical degrees of freedom. The nested-model F-test statistic is
Under and the normal linear model assumptions, . Its 5% upper critical value is , so do not reject the absence of day effects; the p-value is approximately . Thus the missing row has two statistical degrees of freedom, mean square , and the F-test value above. These observations do not provide evidence that including day improves the chocolate-adjusted mean model. They do not prove that every possible day effect or chocolate–day interaction term is absent.
Index chocolate by , day by in the three observed categories, and replicate by . The additive two-factor normal linear model is
Here is the board count, is the mean for the reference chocolate and reference day, and are chocolate and day fixed effects. With corner-point constraints, set and , where is the first level in the day factor. The printed coefficient-free output does not determine that factor ordering; the model is unchanged by a different reference category. There is no chocolate–day interaction term in this fit. The six free mean statistical parameters consist of one baseline, three chocolate contrasts and two day contrasts; the common error variance supplies a further statistical parameter.
Group means are independent random variables with normal distributions
The resulting univariate sampling distributions are
The same variance calculation holds for every , since the difference involves two independent group means. Therefore the standard errors satisfy
Replacing by a common estimated error standard deviation preserves this ratio. Although the individual group means are independent, the contrasts share the baseline mean and have covariance for distinct .
Write . Apart from a constant, the log-likelihood is
For each group,
Thus maximum likelihood estimation sets . The corner-point constraint identifies , giving
The baseline mean is the first group mean, rather than the grand mean, because of the chosen identifiability constraint.
A natural next fit is quadratic regression, which remains a normal linear model in its unknown regression coefficients:
The new design matrix has rows and must have matrix rank three; three distinct predictor values suffice. The curved regression residual pattern suggests trying a positive quadratic term, but its sign and adequacy should be checked after fitting. Inspect the new regression residuals to see whether the systematic curvature has disappeared.
The simple linear regression assumes a straight conditional mean, , so that errors are centred at zero throughout the predictor range. The displayed regression residuals are predominantly positive at both ends and negative in the middle. This is a residual curvature diagnostic: the fitted straight line misses a curved conditional mean. The main concern is the shape of the mean function. The plot alone does not establish failure of the normal distribution assumption or a particular error variance model.
Because both vectors are linear transformations of the same multivariate normal response, is jointly multivariate normal. Its cross-covariance matrix is
Zero cross-covariance implies independence for jointly multivariate normal vectors, including singular ones. Therefore the fitted values and the entire vector of regression residuals are independent. The fitted-residual orthogonality identity gives the zero covariance; the normal distribution assumption is what upgrades it to independence.
The ordinary least squares estimator is . Consequently the fitted values and regression residuals are
An affine transformation of a multivariate normal distribution is again multivariate normal, possibly with a singular covariance matrix. For a random vector with covariance matrix , its transformed covariance matrix is . Since , and , these results give
Both multivariate normal distributions are supported on their respective projected subspaces. In particular, has variance and has variance ; the regression residuals need not be mutually independent.
Full column matrix rank makes a positive-definite matrix, and its matrix inverse is symmetric. Thus the hat matrix satisfies
Also and the image of is contained in the column space of . These identities show that is the orthogonal projection matrix onto the column space of .
Use the normal linear model
Here is the response random vector, is the known design matrix, contains the unknown regression coefficients, and is the common error variance. Conditional on , the errors have normal distributions and are independent random variables. Require for identifiability of and invertibility of . Usually is needed to estimate the error variance from the regression residuals. An intercept, when included, is represented by a column of ones in .
A Steiner system is an -point set together with -element blocks such that every -element subset is contained in exactly one block. On , define the following six-element blocks using the incidence extensions of .
Take the two blocks and . For each duad of and each duad in the syntheme , take
These give forty-five blocks of type and forty-five of type . Finally, for every corresponding partition pair and , take all four unions
There are ten partition pairs and forty blocks of type . Distinct indexing data give distinct blocks within each family, and different types have different intersection sizes with . Thus the total is
To prove the defining property, fix a five-element subset and set .
If or , only or can contain it. If , write and with . A containing block must have type and its omitted duad is . The total has exactly one syntheme containing ; the other total containing that syntheme determines a unique second point . Thus the unique block is .
If , write and . A containing block must have type , with omitted duad . Exactly one duad contains , so is the unique block.
If , write and , where . There are only two possible types. A block exists exactly when the matching has a duad contained in . There is then exactly one such duad, since two disjoint duads cannot fit inside a triple. For a perfect matching on two triples, either all three pairs are cross, or there is one internal pair in each triple and one cross pair. Thus a block exists precisely when is not entirely cross for .
On the other hand, a block containing must use the unique partition of corresponding to . It exists precisely when is contained in one of that partition's triples, and is then unique. By the partition incidence rule in part (b), this happens precisely when is entirely cross. Hence exactly one of the two possible block types exists, always uniquely.
For , interchange the roles of and and use the inverse partition incidence rule proved in part (b). More explicitly, the duad either has an inverse syntheme with an internal pair in the complement of the triple , yielding a unique block, or its inverse syntheme is entirely cross, yielding the unique block. These alternatives are exclusive and exhaustive for the same matching-on-two-triples reason.
Every split therefore gives exactly one containing block. We have constructed
As an independent count, each block contains six five-subsets and . The case proof establishes uniqueness and existence; the count alone would not have done so.

Pinned article: Introduction to the OurBigBook Project

Welcome to the OurBigBook Project! Our goal is to create the perfect publishing platform for STEM subjects, and get university-level students to write the best free STEM tutorials ever.
Everyone is welcome to create an account and play with the site: ourbigbook.com/go/register. We belive that students themselves can write amazing tutorials, but teachers are welcome too. You can write about anything you want, it doesn't have to be STEM or even educational. Silly test content is very welcome and you won't be penalized in any way. Just keep it legal!
We have two killer features:
  1. topics: topics group articles by different users with the same title, e.g. here is the topic for the "Fundamental Theorem of Calculus" ourbigbook.com/go/topic/fundamental-theorem-of-calculus
    Articles of different users are sorted by upvote within each article page. This feature is a bit like:
    • a Wikipedia where each user can have their own version of each article
    • a Q&A website like Stack Overflow, where multiple people can give their views on a given topic, and the best ones are sorted by upvote. Except you don't need to wait for someone to ask first, and any topic goes, no matter how narrow or broad
    This feature makes it possible for readers to find better explanations of any topic created by other writers. And it allows writers to create an explanation in a place that readers might actually find it.
    Figure 1.
    Screenshot of the "Derivative" topic page
    . View it live at: ourbigbook.com/go/topic/derivative
  2. local editing: you can store all your personal knowledge base content locally in a plaintext markup format that can be edited locally and published either:
    This way you can be sure that even if OurBigBook.com were to go down one day (which we have no plans to do as it is quite cheap to host!), your content will still be perfectly readable as a static site.
    Figure 5. . You can also edit articles on the Web editor without installing anything locally.
    Video 3.
    Edit locally and publish demo
    . Source. This shows editing OurBigBook Markup and publishing it using the Visual Studio Code extension.
  3. https://raw.githubusercontent.com/ourbigbook/ourbigbook-media/master/feature/x/hilbert-space-arrow.png
  4. Infinitely deep tables of contents:
    Figure 6.
    Dynamic article tree with infinitely deep table of contents
    .
    Descendant pages can also show up as toplevel e.g.: ourbigbook.com/cirosantilli/chordate-subclade
All our software is open source and hosted at: github.com/ourbigbook/ourbigbook
Further documentation can be found at: docs.ourbigbook.com
Feel free to reach our to us for any help or suggestions: docs.ourbigbook.com/#contact