How to calculate average inter-arrival time from data set?

Hi all,

I am attempting to calculate the average inter-arrival time of hospital admissions for each hour during the afternoon shift (14:00-22:00) for each of the 45 days that I have data (note: the days are not consecutive, they are only weekdays). Ultimately, I would like to save the output in a table for easier viewing if possible. I have attached both a schematic of what I am trying to do and the txt file of data.

I have attempted to use loops to do this but my intution says there is an easier way to do this so I figured I would ask the community. Help would be greatly appreciated as I have never computed anything like this using matlab before.

Matt

Answer 1

[cdawg]

I took a shot at this even though I had for loops :-)

Enjoyed this (even if it might not be correct- haha) thanks!!

%% IMPORT TEXT FILE
opts = delimitedTextImportOptions("NumVariables", 4);

% Specify range and delimiter
opts.DataLines = [2, Inf];
opts.Delimiter = "\t";

% Specify column names and types
opts.VariableNames = ["Diagnosis", "Date", "Time", "Region"];
opts.VariableTypes = ["categorical", "datetime", "datetime", "categorical"];

% Specify file level properties
opts.ExtraColumnsRule = "ignore";
opts.EmptyLineRule = "read";

% Specify variable properties
opts = setvaropts(opts, ["Diagnosis", "Region"], "EmptyFieldRule", "auto");
opts = setvaropts(opts, "Date", "InputFormat", "yyyy-MM-dd");
opts = setvaropts(opts, "Time", "InputFormat", "HH:mm");

% Import the data
data = readtable("pts_admission_Feb_March.txt", opts);

clear opts
data = removevars(data,["Diagnosis","Region"]);

%% START DATA ANALYSIS
% Remove irrelevant data (not within afternoon shift)
data = data(isbetween(data.Time, datetime('14:00','InputFormat','HH:mm'),datetime('22:00','InputFormat','HH:mm')),:);

% Start sorting the data based on date/time
data.Date = categorical(data.Date,unique(data.Date),cellstr(strcat('Day ',string(1:length(unique(data.Date))))));
times = split(string(data.Time),':');
data.Hours = categorical(times(:,1), unique(times(:,1)), cellstr(unique(times(:,1))));
data.mins = str2double(times(:,2));

% Initialize tables
avs = zeros(length(categories(data.Date)), length(categories(data.Hours))-1);
ci = zeros(length(categories(data.Date)), length(categories(data.Hours))-1);
cat1 = categories(data.Date);
cat2 = categories(data.Hours);

z = 1.96; % for 95% CI

for ii = 1:length(cat1)
    datii = data(data.Date == cat1{ii},:);
    for jj = 1:length(cat2)-1
        datjj = datii(datii.Hours == cat2{jj},:);
        n = length(datjj.mins);        
        avs(ii,jj) = mean(diff(datjj.mins));
        s = std(diff(datjj.mins));
        ci(ii,jj) = z*s/sqrt(n);
    end

end


meanTimes = array2table(avs, "VariableNames", strcat('Hour',cat2(1:end-1)), "RowNames",cat1)

meanTimes = 45×8 table
             Hour14    Hour15    Hour16    Hour17    Hour18    Hour19    Hour20    Hour21
             ______    ______    ______    ______    ______    ______    ______    ______

    Day1          4         6    2.5556     5.125         3       3.8       4.6      10.2
    Day2     2.8824       4.5       3.5    3.3571     6.625    5.7778    4.4545       5.3
    Day3     3.8667    4.9091         4    5.8333       3.5    4.7778    4.1429         6
    Day4          3       4.2    4.2857    5.2727       5.6         8       3.6       2.5
    Day5        4.8       3.8    4.5833       5.5    6.1667    4.8571    4.7273        45
    Day6     5.7778      4.75       3.7    6.5556    3.4286    4.7273        10      10.2
    Day7     4.2727    3.8667    4.8333    3.4118       4.3    4.1538         3       4.7
    Day8     2.8235    3.4667    4.6667     13.75    5.6667      5.25    7.7143      5.75
    Day9     3.8571    2.9333    3.8571         4    4.7143       5.7      3.25    6.8333
    Day10    2.4762    3.4667    4.0909    4.0833    3.9231       4.5       5.1       6.5
    Day11    3.8889    2.3636     7.125       9.8       5.1       5.6    4.2727       6.8
    Day12       5.7    4.6667    3.6667       4.2       3.6    3.7333       5.5    2.5556
    Day13       5.6       3.8     3.375    5.1111    7.8333    4.2222       4.2    4.1818
    Day14    3.1176    3.6667    3.8667    8.3333       4.9    7.4286    5.5556    5.8889
    Day15       5.2    3.9231       4.7    4.8182    5.2222    4.1818       5.1    6.4286
    Day16    2.5556       4.5    2.7222       4.6    3.5625    10.667    3.0714      10.8

CI = array2table(ci, "VariableNames", strcat('Hour',cat2(1:end-1)), "RowNames",cat1)

CI = 45×8 table
             Hour14     Hour15     Hour16    Hour17    Hour18    Hour19    Hour20    Hour21 
             _______    _______    ______    ______    ______    ______    ______    _______

    Day1      1.5022     4.3936    1.1969    1.8947    1.0872    1.5737    2.5563     4.3973
    Day2      1.0053     2.6378     1.876    1.6321    3.5774    1.6911     1.831     1.7849
    Day3      1.6447     2.4722    2.3063    3.3268    1.3426    1.8794    4.0688     4.9326
    Day4      1.1644     1.9856    2.6452    1.4556      3.75    7.6987    1.6124    0.80017
    Day5      3.3174      2.022    2.1677    2.4939    4.1753    1.9774    2.8861          0
    Day6      3.8555     3.1793    3.0011    3.1008    1.0637    2.6549    7.2281     9.8228
    Day7       1.541     1.5031    1.8664    1.8627     1.763    1.7744      0.98     2.8689
    Day8      1.1461     1.8784    1.9985    8.1405    2.0084    5.1298    5.3932     2.6308
    Day9      1.8816      1.404    2.0805    1.9114    3.3881    4.4578    2.2398     4.9897
    Day10    0.81049     1.5031    2.2696    1.8754    1.8078    1.3141    3.5616      3.276
    Day11     1.1778    0.76245    2.0194    7.0077    1.8151    3.9417    1.8781     4.4695
    Day12     2.9751     2.7719    2.0844    1.5955     1.408     1.404    2.2156     1.7286
    Day13     2.6894     1.9617    1.6357    4.3985      5.12    2.0031    1.9856     1.6557
    Day14     1.4598     1.5896    1.6131    7.6797    2.5902    4.3801    3.6813     2.4094
    Day15     3.8682     1.5636    1.7849    2.0055    2.6977    2.0372    2.9673     3.2713
    Day16    0.96591     4.0008    1.1927    2.0509    1.9713    7.9816    1.0017     6.7755

So the confidence intervals are meanTimes +/- CI?

[matt beemsterboer]

This code seems to be correct. Thank you for the help, however how would you "refine" the code to do the same exact thing except for each of the four diagnoses (M (mental health), R (respiratory), T (trauma) and C (cardiovascular)?

Given the way you did it, I feel like you would simply add a conditional statement but I am not sure.

Thanks,

Matt