0527127150
120
41
4920
StoneBr
TwnhsE
5
0527145080
120
43
5005
StoneBr
TwnhsE
5
0527146030
120
39
5389
StoneBr
TwnhsE
5
0527162130
060
60
7500
Gilbert
1Fam
5
0527163010
060
75
10000
Gilbert
1Fam
5
0527165230
020
NaN
7980
Gilbert
1Fam
7
0527166040
060
63
8402
Gilbert
1Fam
5
0527180040
020
85
10176
Gilbert
1Fam
5
0527182190
120
NaN
6820
StoneBr
TwnhsE
5
0527216070
060
47
53504
StoneBr
1Fam
5
⋮
All the text data has been converted to categorical variables. But there are still a few things to
clean up.
The OverallCond variable was read in as a numeric array, but its values are all drawn from the
integers 1-10. You can leave these values as numeric data, but you can think of it as ordinal
categorical data. When a categorical array is ordinal, its categories have a specified order. For
example, the categories 10 and 5 can be compared (10 > 5, because a house whose condition is
rated as a 10 is theoretically nicer than one rated 5), but for these comparisons, there is no numeric
meaning to 10 - 5. To avoid unintentionally treating OverallCond as numeric data, convert it to an
ordinal categorical array, which still enables relational comparisons but prevents arithmetic
operations. The category names 1, 2, and so on are easy to interpret and are acceptable.
housing.OverallCond = categorical(housing.OverallCond,1:10,"Ordinal",true);
Similarly, the MSSubClass variable consisted of numeric codes in the original spreadsheet. You can
think of those values as being categorical data. Because there is no mathematical order to these
particular codes, the categories are nonordinal (or nominal). In this case, readtable read those
values in as text to preserve leading zeroes in the codes. MSSubClass was then converted to
categorical data.
While MSSubClass has the data type that you want, you might find it difficult to interpret the codes
as categories of houses. The file that describes the Ames Housing Data contains the definitions of the
numeric codes. Giving these categories readable names can help you understand the data. To make it
clear which names go with which numbers, specify both the categories (code) and their names
(subclass) in another call to the categorical function.
code = ["020" "030" "040" "045" "050" "060" "070" "075" "080" "085" "090" "120" "150" "160" "180"
subclass = ["1-STORY 1946 & NEWER ALL STYLES" ...
"1-STORY 1945 & OLDER" ...
"1-STORY W/FINISHED ATTIC ALL AGES" ...
"1-1/2 STORY - UNFINISHED ALL AGES" ...
"1-1/2 STORY FINISHED ALL AGES" ...
"2-STORY 1946 & NEWER" ...
"2-STORY 1945 & OLDER" ...
"2-1/2 STORY ALL AGES" ...
"SPLIT OR MULTI-LEVEL" ...
"SPLIT FOYER" ...
"DUPLEX - ALL STYLES AND AGES" ...
"1-STORY PUD (Planned Unit Development) - 1946 & NEWER" ...
"1-1/2 STORY PUD - ALL AGES" ...
"2-STORY PUD - 1946 & NEWER" ...
"PUD - MULTILEVEL - INCL SPLIT LEV/FOYER" ...
"2 FAMILY CONVERSION - ALL STYLES AND AGES"];
housing.MSSubClass = categorical(housing.MSSubClass,code,subclass);
9 Tables
9-70
120
41
4920
StoneBr
TwnhsE
5
0527145080
120
43
5005
StoneBr
TwnhsE
5
0527146030
120
39
5389
StoneBr
TwnhsE
5
0527162130
060
60
7500
Gilbert
1Fam
5
0527163010
060
75
10000
Gilbert
1Fam
5
0527165230
020
NaN
7980
Gilbert
1Fam
7
0527166040
060
63
8402
Gilbert
1Fam
5
0527180040
020
85
10176
Gilbert
1Fam
5
0527182190
120
NaN
6820
StoneBr
TwnhsE
5
0527216070
060
47
53504
StoneBr
1Fam
5
⋮
All the text data has been converted to categorical variables. But there are still a few things to
clean up.
The OverallCond variable was read in as a numeric array, but its values are all drawn from the
integers 1-10. You can leave these values as numeric data, but you can think of it as ordinal
categorical data. When a categorical array is ordinal, its categories have a specified order. For
example, the categories 10 and 5 can be compared (10 > 5, because a house whose condition is
rated as a 10 is theoretically nicer than one rated 5), but for these comparisons, there is no numeric
meaning to 10 - 5. To avoid unintentionally treating OverallCond as numeric data, convert it to an
ordinal categorical array, which still enables relational comparisons but prevents arithmetic
operations. The category names 1, 2, and so on are easy to interpret and are acceptable.
housing.OverallCond = categorical(housing.OverallCond,1:10,"Ordinal",true);
Similarly, the MSSubClass variable consisted of numeric codes in the original spreadsheet. You can
think of those values as being categorical data. Because there is no mathematical order to these
particular codes, the categories are nonordinal (or nominal). In this case, readtable read those
values in as text to preserve leading zeroes in the codes. MSSubClass was then converted to
categorical data.
While MSSubClass has the data type that you want, you might find it difficult to interpret the codes
as categories of houses. The file that describes the Ames Housing Data contains the definitions of the
numeric codes. Giving these categories readable names can help you understand the data. To make it
clear which names go with which numbers, specify both the categories (code) and their names
(subclass) in another call to the categorical function.
code = ["020" "030" "040" "045" "050" "060" "070" "075" "080" "085" "090" "120" "150" "160" "180"
subclass = ["1-STORY 1946 & NEWER ALL STYLES" ...
"1-STORY 1945 & OLDER" ...
"1-STORY W/FINISHED ATTIC ALL AGES" ...
"1-1/2 STORY - UNFINISHED ALL AGES" ...
"1-1/2 STORY FINISHED ALL AGES" ...
"2-STORY 1946 & NEWER" ...
"2-STORY 1945 & OLDER" ...
"2-1/2 STORY ALL AGES" ...
"SPLIT OR MULTI-LEVEL" ...
"SPLIT FOYER" ...
"DUPLEX - ALL STYLES AND AGES" ...
"1-STORY PUD (Planned Unit Development) - 1946 & NEWER" ...
"1-1/2 STORY PUD - ALL AGES" ...
"2-STORY PUD - 1946 & NEWER" ...
"PUD - MULTILEVEL - INCL SPLIT LEV/FOYER" ...
"2 FAMILY CONVERSION - ALL STYLES AND AGES"];
housing.MSSubClass = categorical(housing.MSSubClass,code,subclass);
9 Tables
9-70