How to let Matlab fill in different date, string, and numbers in a table

Suppose I have table T1

Date City Vistors
------------------
January 1 2020 NY 100
January 2 2020 NY 110

and table T2

Date City Vistors
------------------
January 4 2020 NY 130
January 5 2020 NY 140

Then I want to have

Date City Vistors
------------------
January 1 2020 NY 100
January 2 2020 NY 110
January 3 2020 NY 120
January 4 2020 NY 130
January 5 2020 NY 140

That is,

(1) I want to vertcat(T1,T2)

(2) Fill in missing date

(3) Fill in missing value for state with the same state name

(4) Fill in missing value for vistors with linear interporation

[dpb]

T=[T1:T2];

See

doc fillmissing

for the rest, is straightforward and examples provided.

Alternatively, you might consider a timetable instead of regular table; then retime could be helpful

[alpedhuez]

No. retime does not work when there are strings in a table.

[Adam Danz]

I'm guessing the city variables are strings. They can be converted to categorical variables. retime handles categoricals (example). If the date-time values are strings, convert them to datetime values.

[Adam Danz]

You mentioned in your duplicate question that you want to use outerjoin. Outerjoin will horizontally concatenate the tables but the example in your question is a vertical concatenation.

[dpb]

What's complicated about

T1.City=categorical(T1.City);

???

Have you even tried anything suggested?

[Adam Danz]

@alpedhuez, you've asked and hopefully learned from 346 questions to date which is amazing. You got this!

[alpedhuez]

Thank you. I am aware of the conversion method. I would hope there is some comparison table among different data types of variables. This support community is definitely the strength of Matlab.

[dpb]

" I would hope there is some comparison table among different data types of variables"

What does that mean? Give a specific example. You can't compare a string to a numeric value directly or the like if that is what you mean, no.

[Adam Danz]

Maybe you could show us what you're trying to compare.

To summarize our suggestions so far,

1. Vertically concatenate the tables assuming they have the same number of columns and same variable names (T=[T1;T2])
2. Convert the City names from string to categorical (see categorical as dpb showed you)

Then you have 2 options,

1. Use fillmissing , or
2. Convert the table to timetable if it isn't already (see table2timetable) and use retime.

If you're stuck at one of those steps or one of these steps won't work with your data, please explain.

[alpedhuez]

I meant a comparison table like

https://www.google.com/search?q=comparison+table+data+type+c%2B%2B&tbm=isch&ved=2ahUKEwjM2r2n-sbtAhXG2VkKHc2gCl0Q2-cCegQIABAA&oq=comparison+table+data+type+c%2B%2B&gs_lcp=CgNpbWcQA1DjQFjBRWCZR2gAcAB4AIABcYgB-gKSAQMzLjGYAQCgAQGqAQtnd3Mtd2l6LWltZ8ABAQ&sclient=img&ei=vu_TX8yDCsaz5wLNwaroBQ&bih=417&biw=910&rlz=1C1GCEA_enUS928US928#imgrc=3PL9t3fxXiwJmM