| Statistics Toolbox™ | |
B = grpstats(A,groupvars)
B = grpstats(A,groupvars,whichstats)
B = grpstats(A,groupvars,whichstats,...,'DataVars',vars)
B = grpstats(A,groupvars,whichstats,...,'VarNames',names)
B = grpstats(A,groupvars) returns a dataset array B that contains the means, computed by group, for variables in the dataset array A. The optional input groupvars specifies the variables in A that define the groups. groupvars can be a positive integer, a vector of positive integers, a variable name, a cell array containing one or more variable names, or a logical vector. groupvars can also be [] or omitted to compute the means of the variables in A without grouping. Grouping variables can be vectors of categorical, logical, or numeric values, a character array of strings, or a cell vector of strings. (See Grouped Data.)
B contains the grouping variables, plus a variable giving the number of observations in A for each group, plus a variable for each of the remaining variables in A. B contains one observation for each group of observations in A.
grpstats treats NaNs as missing values, and removes them.
B = grpstats(A,groupvars,whichstats) returns a dataset array B with variables for each of the statistics specified in whichstats, applied to each of the nongrouping variables in A. whichstats can be a single function handle or name, or a cell array containing multiple function handles or names. The names can be chosen from among the following:
'mean' — mean
'sem' — standard error of the mean
'numel' — count, or number of non-NaN elements
'gname' — group name
'std' — standard deviation
'var' — variance
'meanci' — 95% confidence interval for the mean
'predci' — 95% prediction interval for a new observation
Each function included in whichstats must accept a subset of the rows of a dataset variable, and compute column-wise descriptive statistics for it. A function should typically return a value that has one row but is otherwise the same size as its input data. For example, @median and @skewness are suitable functions to apply to a numeric dataset variable.
A summary statistic function may also return values with more than one row, provided the return values have the same number of rows each time grpstats applies the function to different subsets of data from a given dataset variable. For a dataset variable that is nobs-by-m-by-... if a summary statistic function returns values that are nvals-by-m-by-... then the corresponding summary statistic variable in B is ngroups-by-m-by-...-by-nvals, where ngroups is the number of groups in A.
B = grpstats(A,groupvars,whichstats,...,'DataVars',vars) specifies the variables in A to which the functions in whichstats should be applied. The output dataset arrays contain one summary statistic variable for each of the specified variables. vars is a positive integer, a vector of positive integers, a variable name, a cell array containing one or more variable names, or a logical vector.
B = grpstats(A,groupvars,whichstats,...,'VarNames',names) specifies the names of the variables in B. By default, grpstats uses the names from A for the grouping variables, and constructs names for the summary statistic variables based on the function name and the data variable names. The number of variables in B is ngroupvars + 1 + ndatavars*nfuns, where ngroupvars is the number of variables specified in groupvars, ndatavars is the number of variables specified in vars, and nfuns is the number of summary statistics specified in whichstats.
Compute blood pressure statistics for the data in hospital.mat, by sex and smoker status:
load hospital
grpstats(hospital,...
{'Sex','Smoker'},...
{@median,@iqr},...
'DataVars','BloodPressure')
ans =
Sex Smoker GroupCount
Female_0 Female false 40
Female_1 Female true 13
Male_0 Male false 26
Male_1 Male true 21
median_BloodPressure
Female_0 119.5 79
Female_1 129 91
Male_0 119 79
Male_1 129 92
iqr_BloodPressure
Female_0 6.5 5.5
Female_1 8 5.5
Male_0 7 6
Male_1 10.5 4.5 | grpstats | gscatter | |
| © 1984-2008- The MathWorks, Inc. - Site Help - Patents - Trademarks - Privacy Policy - Preventing Piracy - RSS |