1 Introduction / notes

Index fitting report for reference grid OMs with length composition weighting of 1/20 (factor level L)

## R Version:  R version 4.0.4 (2021-02-15)

2 Residuals in indices

3 Standard Deviation of Residuals

Table 1. Standard deviation in log residuals

4 Standard Error of Residuals

Table 2. Standard error in log residuals (St.Dev/(n)^0.5)

5 Autocorrelation of Residuals

Table 3. Lag-1 autocorrelation in log residuals

6 Runs in residuals: proportion crosses

Table 4. Proportion crosses (crosses / (n-1))

7 Runs in residuals: runs statistic

Table 5. Runs statistic (p-value)

8 Maximum length of recent runs

Table 6. Maximum run (positive or negative residuals) in last 10 years

9 Mean statistics (all OMs)

Table 7a. Summary statistics for the West Area indices (mean values) over all operating models. A.C. = lag-1 temporal autocorrelation, Prop. Crs = number of residuals divided by length of index - 1. Max run 10 = maximum size of a positive or negative run over the last 10 years. Runs p = the p-value of the runs statistic.

Figure 2. West area indices (red) and OM fits (blue).

Table 7b. As Table 7a but for the East area indices.

Figure 3. East area indices (red) and OM fits (blue).

10 Uncertainty in standard deviation (all OMs)

Table 8a. Uncertainty in standard deviation in log residuals of West area indices (generated by an inverse chi-squared distribution) expressed by the 5th, 25th, 50th, 75th and 95th percentiles.

Table 8b. Uncertainty in standard deviation in log residuals of East area indices (generated by an inverse chi-squared distribution) expressed by the 5th, 25th, 50th, 75th and 95th percentiles.

11 Interquantile ranges of SD estimates

Table 9a. The 25th percentile of the sampled St.Dev by index and operating model

Table 9b. The 75th percentile of the sampled St.Dev by index and operating model

Figure 4. The interquartile range of SD estimates by index