MCHART Statement: SHEWHART Procedure

Constructing Median Charts

The following notation is used in this section:

$\mu $

process mean (expected value of the population of measurements)

$\sigma $

process standard deviation (standard deviation of the population of measurements)

$\bar{X}_{i}$

mean of measurements in ith subgroup

$n_{i}$

sample size of ith subgroup

N

the number of subgroups

$x_{ij}$

jth measurement in the ith subgroup, $j=1,2,3, \ldots , n_{i}$

$x_{i(j)}$

jth largest measurement in the ith subgroup. Then

\[ x_{i(1)} \leq x_{i(2)} \leq \ldots \leq x_{i(n_{i})} \]

$\overline{\overline{X}}$

weighted average of subgroup means

$M_{i}$

median of the measurements in the ith subgroup:

\[ M_ i = \left\{ \begin{array}{ll} x_{i((n_ i + 1)/2)} & \mbox{if $n_{i}$ is odd} \\ (x_{i(n_ i/2)} + x_{i((n_ i/2)+1)})/2 & \mbox{if $n_{i}$ is even} \end{array} \right. \]

$\bar{M}$

average of the subgroup medians:

\[ \bar{M}=(n_1M_1+\ldots +n_ NM_ N)/(n_1+\ldots +n_ N) \]

$\tilde{M}$

median of the subgroup medians. Denote the jth largest median by $M_{(j)}$ so that $M_{(1)} \leq M_{(2)}\leq \ldots \leq M_{(N)}$. Then

\[ \tilde{M} = \left\{ \begin{array}{ll} M_{((N+1)/2)} & \mbox{if } N \mbox{ is odd} \\ (M_{(N/2)} + M_{(N/2)+1})/2 & \mbox{if } N \mbox{ is even} \end{array} \right. \]

$e_ M(n)$

standard error of the median of n independent, normally distributed variables with unit standard deviation (the value of $e_ M(n)$ can be calculated with the STDMED function in a DATA step)

$Q_{p}(n)$

100pth percentile (0 < p < 1) of the distribution of the median of n independent observations from a normal population with unit standard deviation

$z_{p}$

100pth percentile of the standard normal distribution

$D_{p}(n)$

100pth percentile of the distribution of the range of n independent observations from a normal population with unit standard deviation

Plotted Points

Each point on a median chart indicates the value of a subgroup median ($M_{i}$). For example, if the tenth subgroup contains the values 12, 15, 19, 16, and 14, the value plotted for this subgroup is $M_{10} = 15$.

Central Line

The value of the central line indicates an estimate for $\mu $, which is computed as

  • $\bar{M}$ by default

  • $\overline{\overline{X}}$ when you specify MEDCENTRAL=AVGMEAN

  • $\tilde{M}$ when you specify MEDCENTRAL=MEDMED

  • $\mu _{0}$ when you specify $\mu _0$ with the MU0= option

Control Limits

You can compute the limits

  • as a specified multiple (k) of the standard error of $M_{i}$ above and below the central line. The default limits are computed with k = 3 (these are referred to as $3\sigma $ limits).

  • as probability limits defined in terms of $\alpha $, a specified probability that $M_{i}$ exceeds the limits

The following table provides the formulas for the limits:

Table 17.26: Limits for Median Charts

Control Limits

LCLM = lower limit = $\bar{M} - k\hat{\sigma }e_{M}(n_ i)$

UCLM = upper limit = $\bar{M} + k\hat{\sigma }e_{M}(n_ i)$

Probability Limits

LCLM = lower limit = $\bar{M} - Q_{\alpha /2}(n_ i)\hat{\sigma }$

UCLM = upper limit = $\bar{M} + Q_{1-\alpha /2}(n_ i)\hat{\sigma }$


Note that the limits vary with $n_{i}$. In Table 17.26, replace $\bar{M}$ with $\overline{\overline{X}}$ if you specify MEDCENTRAL=AVGMEAN, and replace $\bar{M}$ with $\tilde{M}$ if you specify MEDCENTRAL=MEDMED. Replace $\bar{M}$ with $\mu _0$ if you specify $\mu _0$ with the MU0= option, and replace $\hat{\sigma }$ with $\sigma _0$ if you specify $\sigma _0$ with the SIGMA0= option. The formulas assume that the data are normally distributed.

You can specify parameters for the limits as follows:

  • Specify k with the SIGMAS= option or with the variable _SIGMAS_ in a LIMITS= data set.

  • Specify $\alpha $ with the ALPHA= option or with the variable _ALPHA_ in a LIMITS= data set.

  • Specify a constant nominal sample size $n_{i} \equiv n$ for the control limits with the LIMITN= option or with the variable _LIMITN_ in a LIMITS= data set.

  • Specify $\mu _{0}$ with the MU0= option or with the variable _MEAN_ in the LIMITS= data set.

  • Specify $\sigma _{0}$ with the SIGMA0= option or with the variable _STDDEV_ in the LIMITS= data set.