How can i filter a table with several variables

Hey guys,

i have a table with four rows: String, String, Datetime and Double. I want to generate a new table with all four rows filtered by a startdate and a enddate that i can analyse for example just 10 years (2000-2010).

Thank you :)

[MarKf]

%% sample data
dates = datetime - years(5:5:40)' + hours(randn(8,1));
doublesx = randn(8,1);
strings1 = ('a':char('a'+7))';
strings2 = ('m':char('m'+7))';
example_table = table(strings1,strings2,dates,doublesx)

example_table = 8×4 table
    strings1    strings2           dates            doublesx
    ________    ________    ____________________    ________

       a           m        03-Jan-2018 10:09:18    -0.92048
       b           n        03-Jan-2013 04:37:03     0.25715
       c           o        04-Jan-2008 00:52:16     0.79814
       d           p        03-Jan-2003 18:22:47    -0.26506
       e           q        03-Jan-1998 14:05:05      1.3091
       f           r        03-Jan-1993 11:11:26     -1.5165
       g           s        04-Jan-1988 04:03:51     0.74112
       h           t        03-Jan-1983 23:47:08     -1.3076

%% new table selct
idx = (example_table.dates.Year >= 2000) & (example_table.dates.Year < 2010);
new_table = example_table(idx,:) %even if 3rd position dates

new_table = 2×4 table
    strings1    strings2           dates            doublesx
    ________    ________    ____________________    ________

       c           o        04-Jan-2008 00:52:16     0.79814
       d           p        03-Jan-2003 18:22:47    -0.26506

P.S. Actual or faux example table .mat file or generating code shared would be appreaciated and make it easier to help. You likely mean for example 4 colums and many row entries, but in case you do have a horizontal table it'd work all the same. Also the number of variables does not matter. You could also convert it to a timetable using table2timetable and then use timerange

Answer 1

[Voss]

% an example table with four columns:
dt = datetime(2023,1,(1:10).');
A = string(rand(10,1));
B = string(rand(10,1));
C = rand(10,1);
t = table(A,B,dt,C)

t = 10×4 table
        A             B             dt            C    
    __________    __________    ___________    ________

    "0.96935"     "0.017237"    01-Jan-2023     0.42444
    "0.92083"     "0.17279"     02-Jan-2023     0.39157
    "0.054817"    "0.93052"     03-Jan-2023      0.1114
    "0.21611"     "0.74939"     04-Jan-2023      0.1513
    "0.36615"     "0.32896"     05-Jan-2023     0.51285
    "0.56422"     "0.20432"     06-Jan-2023     0.68134
    "0.53045"     "0.92401"     07-Jan-2023    0.062899
    "0.84537"     "0.70987"     08-Jan-2023     0.35308
    "0.18631"     "0.23875"     09-Jan-2023     0.71427
    "0.13705"     "0.38498"     10-Jan-2023     0.34559

% a logical array to be used as a filter (keep data from dates between 
% Jan 3, 2023 and Jan 7, 2023, inclusive):
startDate = datetime(2023,1,3);
endDate = datetime(2023,1,7);
idx = t{:,3} >= startDate & t{:,3} <= endDate

idx = 10×1 logical array
   0
   0
   1
   1
   1
   1
   1
   0
   0
   0

% create the new table from just those dates:
new_t = t(idx,:)

new_t = 5×4 table
        A             B            dt            C    
    __________    _________    ___________    ________

    "0.054817"    "0.93052"    03-Jan-2023      0.1114
    "0.21611"     "0.74939"    04-Jan-2023      0.1513
    "0.36615"     "0.32896"    05-Jan-2023     0.51285
    "0.56422"     "0.20432"    06-Jan-2023     0.68134
    "0.53045"     "0.92401"    07-Jan-2023    0.062899

Answer 2

[Steven Lord]

If you turned your table into a timetable using table2timetable you could use a timerange to index into the rows of that timetable.