3
$\begingroup$

Background: I am studying a course on statistical experiments using the textbook by Douglas Montgomery on the analysis of experiments. This is an introductory course and so I am relatively new to the study of these types of models.

Question: Consider the model $Y_{i,j,k} = \mu + R_i + C_j + T_k + \epsilon_{i,j,k}$ where $R_i$ is the impact of from row $i$, $C_j$ is the impact from column $j$ and $T_k$ is the impact from treatment $k$ where there are exactly $n$ rows, $n$ columns, and $n$ treatments. We know that the row sum of squares $S^2_{\text{row}} = \sum _ {i=1}^n \frac{Y_{i \cdot \cdot}^2}{n} - \frac{Y_{\cdot \cdot \cdot}^2}{n^2}$ and similarly for $S^2_{\text{column}}$ and $S^2_{\text{treatment}}$ we have the same form but instead we sum over $j$ and $k$ respectively from $1$ to $n$ (we take the formulae for the sums of squares directly from the corresponding ANOVA table).

I have read in the textbook (without justification) the formula for the expected value $ \color{red}{\mathbb{E}(S^2_{\text{row}}\space)}$ and (again) the same applies for the columns and treatments by replacing $R_i^2$ with $C_j^2$ (for the columns) and $T_k^2$ (for the treatments).

$\text{How do we find the formula for the expectation?}$

Note: we are making the standard modelling assumptions for these types of models where the error terms are independent and normally distributed with mean $0$ and variance $\sigma^2$.

$\endgroup$
11
  • $\begingroup$ Plug $Y_{i..}$ and $Y_{...}$ into $S_{\text{row}}^2$ and do the algebra. $\endgroup$
    – Zhanxiong
    Commented Feb 17, 2023 at 4:27
  • 1
    $\begingroup$ It's easier to work with the centered form of $S_R^2$ (compared with my first comment, although it should also work, but more verbosely), as User1865345's answer shows. In DoE, using orthogonality between effects always helps. $\endgroup$
    – Zhanxiong
    Commented Feb 17, 2023 at 5:10
  • 2
    $\begingroup$ By "orthogonality between effects", I mean all the terms inside parentheses of the decomposition $y_{ijk} - \bar{y}_{...} = (\bar{y}_{i..} - \bar{y}_{...}) + (\bar{y}_{.j.} - \bar{y}_{...}) + (\bar{y}_{..k} - \bar{y}_{...}) + (y_{ijk} - \bar{y}_{i..} - \bar{y}_{.j.} - \bar{y}_{..k} + 2\bar{y}_{...})$ are orthogonal. Using this fact, it is easy to derive all the sum of squares in the ANOVA table and compute expected mean of sum squares. $\endgroup$
    – Zhanxiong
    Commented Feb 17, 2023 at 5:20
  • $\begingroup$ Could you clarify the meaning of the bar notation? I'm not quite following this explanation or the answer as the notation seems to be slightly different from that which I'm familiar with @Zhanxiong $\endgroup$
    – FD_bfa
    Commented Feb 17, 2023 at 5:24
  • 1
    $\begingroup$ $n^3$ is for CRD, not for Latin square. For Latin square, it is $n^2$. Please read your textbook more carefully and ponder the model structure for more time before jumping to math details. $\endgroup$
    – Zhanxiong
    Commented Feb 17, 2023 at 5:47

2 Answers 2

4
$\begingroup$

If you are reading the textbook Design and Analysis of Experiments (8th edition) by Douglas Montgomery, although the model for the Latin squares design is formally written as (note that the notations in the cited textbook were kept unchanged in equation $(1)$. It is obvious that $\alpha, \beta, \tau, p$ in $(1)$ correspond to $R, C, T, n$ in your question): \begin{align} y_{ijk} = \mu + \alpha_i + \tau_j + \beta_k + \epsilon_{ijk} \begin{cases} i = 1, 2, \ldots, p \\ j = 1, 2, \ldots, p \\ k = 1, 2, \ldots, p \end{cases}, \tag{1} \end{align} which seems no different from the main-effects only model for the three-factor factorial design (admittedly, this representation could have been made less ambiguous$^\dagger$), there is an important note below this equation:

Because there is only one observation in each cell, only two of the three subscripts $i, j, k$ are needed to denote a particular observation. This is a consequence of each treatment appearing exactly once in each row and column.

Therefore, instead of interpreting $y_{i..}$ as $\sum_{j = 1}^p\sum_{k = 1}^py_{ijk}$, it should be interpreted as row sum of the $i$-th row, meaning it is the sum of $p$ summands rather than $p^2$. Keeping this in mind, as well as the constraints $\alpha_. = \beta_. = \tau_. = 0$, model $(1)$ entails:

\begin{align} & y_{i..} = p\mu + p\alpha_i + \tau_. + \beta_. + \epsilon_{i..} = p\mu + p\alpha_i + \epsilon_{i..}, \\ & y_{...} = \sum_{i = 1}^p y_{i..} = p^2\mu + p\alpha_. + \epsilon_{...} = p^2\mu + \epsilon_{...}. \tag{2} \end{align}

Since $\epsilon_{i..} \sim N(0, p\sigma^2), \epsilon_{...} \sim N(0, p^2\sigma^2)$, it follows that $E[\epsilon_{i..}^2] = p\sigma^2$, $E[\epsilon_{...}^2] = p^2\sigma^2$, hence $(2)$ implies that (all the cross-product terms between the random error and the fixed effect have expected value $0$ because $E[\epsilon_{i..}] = E[\epsilon_{...}] = 0$): \begin{align} & E(y_{i..}^2) = p^2\mu^2 + p^2\alpha_i^2 + 2p^2\mu\alpha_i + p\sigma^2, \\ & E(y_{...}^2) = p^4\mu^2 + p^2\sigma^2. \tag{3} \end{align}

Substitute $(3)$ back to $E[SS_{\text{rows}}] = \frac{1}{p}\sum_{i = 1}^pE[y_{i..}^2] - \frac{1}{p^2}E[y_{...}^2]$ and use $\alpha_. = 0$, it follows that \begin{align} & E[SS_{\text{rows}}] = \frac{1}{p}\sum_{i = 1}^p(p^2\mu^2 + p^2\alpha_i^2 + 2p^2\mu\alpha_i + p\sigma^2) - \frac{1}{p^2}(p^4\mu^2 + p^2\sigma^2) \\ =& p\sum_{i = 1}^p\alpha_i^2 + p\sigma^2 - \sigma^2 \\ =& (p - 1)\sigma^2 + p\sum_{i = 1}^p\alpha_i^2, \end{align} as desired.


$^\dagger$ For example, a better representation can be: \begin{align} y_{ik} = \mu + \alpha_i + \tau_{L(i, k)} + \beta_k + \epsilon_{ik} \begin{cases} i = 1, 2, \ldots, p \\ k = 1, 2, \ldots, p \end{cases}, \end{align} where $L$ is the Latin square (which is an order $k$ matrix) selected for the experiment. See also this link.

$\endgroup$
5
  • $\begingroup$ That's a straight forward derivation (+1). And yes, Montgomery could have adhered to a better representation of the model to emphasize the overindexing as done by, say, Christensen in his book. $\endgroup$ Commented Feb 17, 2023 at 18:26
  • $\begingroup$ @User1865345 You can provide more detailed information of this book (title, page number) for interested readers. $\endgroup$
    – Zhanxiong
    Commented Feb 17, 2023 at 18:48
  • $\begingroup$ Thanks. I assume $SS_{\text{rows}}$ corresponds to $S_{\text{rows}}$ in the question? $\endgroup$
    – FD_bfa
    Commented Feb 18, 2023 at 13:45
  • $\begingroup$ Yes, I kept all the notations to be consistent with the book I mentioned. $\endgroup$
    – Zhanxiong
    Commented Feb 18, 2023 at 14:45
  • $\begingroup$ Thanks again for your answer (+1). I’ve just had a chance to properly go through it and it all makes sense to me. I’m probably just missing some straight forward ideas with the other answer, but this one has been easier to understand from my perspective $\endgroup$
    – FD_bfa
    Commented Feb 19, 2023 at 4:08
4
$\begingroup$

The model you described is LSD.

Here $$ S^2_R = n\sum_i\left(\bar{y}_{i\cdot\cdot}- \bar{y}_{\cdot\cdot\cdot} \right)^2.$$

Then

\begin{align}\mathbb E\left[S^2_R\right] &= \mathbb E\left[n\sum_i\left(R_i + \bar{\varepsilon}_{i\cdot\cdot} - \bar{\varepsilon}_{\cdot\cdot\cdot}\right)^2\right]\\ &= n\left[\sum_iR_i^2+\mathbb E\left\{\sum_i(\bar{\varepsilon}_{i\cdot\cdot} - \bar{\varepsilon}_{\cdot\cdot\cdot})^2\right\} + 2\underbrace{\sum_i\alpha_i\mathbb E(\bar{\varepsilon}_{i\cdot\cdot} - \bar{\varepsilon}_{\cdot\cdot\cdot})}_{=0}\right]\\ &= n\left[\sum_iR_i^2+\mathbb E\left\{\sum_i\bar{\varepsilon}^2_{i\cdot\cdot} - n\cdot \bar{\varepsilon}^2_{\cdot\cdot\cdot}\right\} \right] \\ &= n\left[\sum_iR_i^2 + \sum_i\frac{\sigma^2} n - \frac{\sigma^2}n\right] \\ &= n\sum_i R_i^2 + (n-1)\sigma^2.\end{align}

$\endgroup$
26
  • $\begingroup$ The notations are universal; so I didn't elaborate them. But if needed, would do that. $\endgroup$ Commented Feb 17, 2023 at 5:07
  • $\begingroup$ Where did this step at the beginning come from? $\mathbb E(S^2_R) = \mathbb E(n\sum_i (R_i + \bar{\varepsilon}_{i\cdot\cdot} - \bar{\varepsilon}_{\cdot\cdot\cdot})^2)$ $\endgroup$
    – FD_bfa
    Commented Feb 17, 2023 at 5:12
  • 1
    $\begingroup$ Your last comment echoed my comment under the question. Do not mechanically do math before really understanding what Latin square design is! The $j, k$ under the sum sign do not vary independently (again, that was for CRD). $\endgroup$
    – Zhanxiong
    Commented Feb 17, 2023 at 6:19
  • 1
    $\begingroup$ You wrote it almost correctly but in place of $\mu^2, $ it would have been $n^2\mu$ and keep in mind that $\sum R_i=\sum C_j=\sum T_k=0.$ $\endgroup$ Commented Feb 17, 2023 at 11:38
  • 1
    $\begingroup$ Thanks for all your help $\endgroup$
    – FD_bfa
    Commented Feb 17, 2023 at 11:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.