Convert date data in datetime
Show older comments
Dear Matlab Community,
I have uplaoded the small subset of the data which I am using;
I am converting it in dateime using:
date = datetime(string(data.Date),'InputFormat','yyyyMMddHH' ) ;
I have also attached the output file.
I will also like to add that the original file is in csv format, I am using 'readtable' to read and process it in Matlab.
The problem is date corresponding to 24hrs is not coverted in datetime.
Any help will be highly appreciated.
Many thanks.
5 Comments
the cyclist
on 19 Dec 2023
Rather than pasting the data (which is potentially ambiguous as to how it is stored), can you upload the data in a MAT file? You can use the paper clip icon in the INSERT section of the toolbar.
Here is a simpler, self-contained example that illustrates @PS's question. The last element does not convert (because the hour is listed as '24'):
date_data = ["2016010101","2016010113","2016010124"];
datetime(date_data,'InputFormat','yyyyMMddHH')
PS
on 19 Dec 2023
Stephen23
on 19 Dec 2023
"I will also like to add that the original file is in csv format, I am using 'readtable' to read and process it in Matlab."
Please upload a sample CSV file, to see if the importing can be improved.
Accepted Answer
Dyuman Joshi
on 19 Dec 2023
Moved: Dyuman Joshi
on 19 Dec 2023
"The problem is date corresponding to 24hrs is not coverted in datetime."
Yes, because in 24 hour format, the hours are counted from 0 to 23.
A work around which works for the data in the given format -
load('data.mat')
format longg
date = datetime(string(data.Date - 1),'InputFormat','yyyyMMddHH')
4 Comments
PS
on 19 Dec 2023
Dyuman Joshi
on 19 Dec 2023
But there are only 49 elements in the array above.
Also, (I guess) you want to display the hours in 1-24 range, which is not possible in MATLAB datetime arrays.
Fangjun Jiang
on 19 Dec 2023
If the data.Date in data.mat is the best format that you prefer, you could apply the same "duration" idea.
date = datetime(string(data.Date - 1),'InputFormat','yyyyMMddHH')+hours(1)
PS
on 20 Dec 2023
More Answers (2)
Fangjun Jiang
on 19 Dec 2023
Edited: Fangjun Jiang
on 19 Dec 2023
You could treat the last two digits as duration. Code below is not optimized but gives the correct result.
load Data.mat;
date_data=char(string(data.Date));
a=char(date_data);
b=a(:,1:end-2);
c=a(:,end-1:end);
d=datetime(b,'InputFormat','yyyyMMdd')+hours(str2num(c))
02-Jan-2016 20:00:00
02-Jan-2016 21:00:00
02-Jan-2016 22:00:00
02-Jan-2016 23:00:00
03-Jan-2016 00:00:00
03-Jan-2016 01:00:00
1 Comment
Another solution is the numeric datetime constructor syntax (i.e. t = datetime(Y,M,D,H,MI,S)), which allows values outside the conventional range:
"If an element of the Y, M, D, H, MI, or S inputs falls outside the conventional range, then datetime adjusts both that element and the same element of the previous input."
Since your timestamps always have the same field widths, you can 1) extract each field, 2) convert to numeric, and 3) call the datetime constructor.
load data.mat
% Convert Date to string
data = convertvars(data, "Date", "string");
% Extract date and time fields
data = addvars(data, extractBetween(data.Date, 1, 4), Before=1, NewVariableNames="Year");
data = addvars(data, extractBetween(data.Date, 5, 6), After=1, NewVariableNames="Month");
data = addvars(data, extractBetween(data.Date, 7, 8), After=2, NewVariableNames="Day");
data = addvars(data, extractBetween(data.Date, 9, 10), After=3, NewVariableNames="Hour");
% Convert fields to numeric
data = convertvars(data, ["Year","Month","Day","Hour"], "double");
% Construct datetime
data = addvars(data, datetime(data.Year, data.Month, data.Day, data.Hour, 0, 0), Before=1, NewVariableNames="Datetime");
tail(data)
Categories
Find more on Dates and Time 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!
