91Ó°ÊÓ

In statistics, the median is a measure of central tendency. It identifies the midpoint of a dataset. To find the median, you first arrange the numbers in ascending order and then pinpoint the middle value(s). If the number of observations is odd, the median is simply the middle number. For even numbers of observations, calculate the median by averaging the two middle numbers.

In our exercise, we start with 12 data points:

• First, sort the data: \[-2.7, -0.8, 0.8, 3.6, 7.1, 8.3, 8.4, 8.6, 9.4, 9.7, 11.4, 18.2\]
• The two middle numbers are 8.3 and 8.4, because the set has 12 numbers (an even set).
• Calculate the median: \[ \text{Median} = \frac{8.3 + 8.4}{2} = 8.35 \]

The median value, 8.35, divides the dataset in half, helping us to differentiate between higher and lower values.

A randomness test, in the context of statistics, helps determine if data points arise in a random order or show a specific pattern. For this, one commonly uses the Runs Test, which assesses sequences of symbols to establish evidence for randomness.

Randomness implies that there is no discernible pattern in the arrangement of data points around the median. This characteristic is essential in various fields for ensuring unbiased results.

In our task, the randomness test uses symbols 'A' and 'B'. This symbolic representation allows us to convert numeric data into categories, making it easier to spot patterns or test randomness.

Symbolic representation involves translating numerical data into simple symbols like 'A' and 'B'. This technique is especially useful when examining data in terms of its relationship to some central value, such as the median.

In this exercise,

• Assign the symbol 'A' for numbers above the median (8.35).
• Assign 'B' for numbers below the median.

The transformation makes the non-numeric sequence: (B, A, B, A, B, A, A, B, A, A, B, B). A symbolic approach emphasizes the movement or 'run' of data around the central point, simplifying further analysis, such as the Runs Test.

The Run Test is a non-parametric test that assesses the sequence of data, specifically the order in which symbols appear. The goal is to identify 'runs'—consecutive sequences of identical symbols—and determine if their occurrence is consistent with randomness. A run is a sequence of similar items, like consecutive 'A's or 'B's in symbolic data.

To perform the Run Test:

• Identify sequential groupings or runs.
• Count the number of these runs.

In our case, there are nine runs: B, A, B, A, B, AA, B, AA, BB.

The expected number of runs \[ E(R) \] is calculated using \[ E(R) = \frac{2n_1n_2}{n} + 1 \], where \[ n_1 \] and \[ n_2 \] are the numbers of A's and B's, respectively. Compute the standard deviation, \[ \sigma(R) \], and the test statistic, \[ Z \], for further analysis. Finally, compare the observed value \[ Z \] against critical values to determine if there's evidence against randomness.