How to sort by time

1 view (last 30 days)
Kanakaiah Jakkula
Kanakaiah Jakkula on 16 Sep 2017
Edited: dpb on 16 Sep 2017
Hi,
I have below data,I want to select two data points above and below to the specified time.
2015/8/5 12:13 2
2015/6/13 22:13 5
2015/9/16 15:45 8
2015/7/23 17:05 15
2015/9/14 12:45 20
2015/6/17 21:44 26
2015/5/13 12:45 28
given time is: 2015/7/23 17:05, two data counts above and below which are more close to give time are"
My desired output:
2015/8/5 12:13 2
2015/6/13 22:13 5
2015/9/14 12:45 20
2015/6/17 21:44 26
many many thanks in advance,

Answers (2)

Andrei Bobrov
Andrei Bobrov on 16 Sep 2017
Edited: Andrei Bobrov on 16 Sep 2017
T = readtable('20170916.txt','CollectOutput',1);
T.Var1 = strcat(T{:,1}(:,1),{' '},T{:,1}(:,2));
T.
t = datetime('2015/7/23 17:05','f','uuuu/MM/dd HH:mm')
z = abs(day(T{:,1} - t))
t = ~z;
y = ~t.*(cumsum(t) + 1 );
n = numel(z);
kk = (1:n)';
s = false(n,1);
for ii = 1:2
m = y == ii;
p = kk(m);
[~,v] = sort(z(m));
s(p(v(1:2))) = true;
end
out = T(s,:)
out =
4×2 table
Date data
________________ ____
2015/08/05 12:13 2
2015/06/13 22:13 5
2015/09/14 12:45 20
2015/06/17 21:44 26

dpb
dpb on 16 Sep 2017
Edited: dpb on 16 Sep 2017
>> dnm=datetime(c(:,1),'format','yyyy/M/d HH:mm'); % convert date strings
>> target=datetime('2015/7/23 17:05','format','yyyy/M/d HH:mm'); % and the target date
>> [dnm,ix]=sort(dnm); % order, keep index
>> idx=[find(dnm<target,2,'last');find(dnm>target,2,'first')] % find closest two either side of target
idx =
2
3
5
6
>> c(ix(idx),:) % look up originals based on presort index
ans =
'2015/6/13 22:13' [ 5]
'2015/6/17 21:44' [26]
'2015/8/5 12:13' [ 2]
'2015/9/14 12:45' [20]
>>
Unlike the previous Answer this replaces, this doesn't assume the target time is identically one of those in the data vector (although that might be always so, this is more generic).
Also the previous note on sort wasn't so; not sure what was wrong that caused that symptom earlier...

Categories

Find more on Linear and Nonlinear Regression in Help Center and File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!