StackOverflow2013

Note that there are some explanatory texts on larger screens.

plurals

PONAs occuring in the output of logistic regression (using multinom) in R
primarykey
Id
19971884
data
AcceptedAnswerId
19974977
AnswerCount
1
ClosedDate
CommentCount
0
CommunityOwnedDate
CreationDate
2013-11-14T07:24:09.037
FavoriteCount
0
LastActivityDate
2013-11-14T10:15:33.287
LastEditDate
LastEditorUserId
0
OwnerUserId
2787586
ParentId
0
PostTypeId
1
Score
0
ViewCount
733
LastEditorDisplayName
text
Body
I'm trying to perform logistic regression using <code>multinom()</code> of <code>nnet</code> package over the following data using R: <a href="https://drive.google.com/file/d/0Bznk6HkWGRBYZ3ZIZzNrdE11TjQ/edit?usp=sharing" rel="nofollow" title="train_data.csv">Train Data</a> <a href="https://drive.google.com/file/d/0Bznk6HkWGRBYczNXeFl5cGxMNGc/edit?usp=sharing" rel="nofollow" title="test_data.csv">Test Data</a> <h3>Data Description:</h3> Each of the train and test files have 10,000 rows of data and 144 variables, where the last column is "Predict", which is the column that should be predicted in the test data. Since, predict contains only two values 0 and 1 (binary), I started with fitting the train data to a logistic regression model using <code>multinom()</code>. Most of the rows are continuous numeric values, while a few fields contains name fields like names of a weekdays ("Sunday", "Monday", ...), which are to be considered as factors. This is the following code used: <pre><code>train_data <- read.csv("train_data.csv") test_data <- read.csv("test_data.csv") library(nnet) my_multinom <- multinom(formula= as.factor(Predict) ~ ., data= train_data) predictions <- predict(my_multinom, test_data[,1:143]) </code></pre> And the output is: <pre><code>> summary(predictions) 0 1 NA's 130 52 9818 </code></pre> where as, there are no <code>NA's</code> in the <code>Predict</code> column of the <code>train_data</code>: <pre><code>> summary(as.factor(train_data$Predict)) 0 1 9734 266 </code></pre> I want to know what is the reason for the occurrence of such a huge amount of <code>NA's</code> (more than 98%) and how to proceed further so that I could avoid these <code>NA's</code>?
Tags
<r><logistic-regression>
Title
NAs occuring in the output of logistic regression (using multinom) in R
singulars
PostAcceptedAnswerId
1. PO
 singulars
 PostTypePostTypeId
 PTAnswer
PostParentId
1. This table or related slice is empty.
PostTypePostTypeId
1. PTQuestion
UserLastEditorUserId
1. This table or related slice is empty.
UserOwnerUserId
1. USStrikeR
plurals
PostLinksPostIdRelatedPostId
1. This table or related slice is empty.
PostLinksRelatedPostIdPostId
1. This table or related slice is empty.
PostsAcceptedAnswerId
1. This table or related slice is empty.
PostsParentIdCreationDate
1. PO
 singulars
 PostTypePostTypeId
 PTAnswer
VotesPostIdCreationDate
1. This table or related slice is empty.
CommentsPostId
1. This table or related slice is empty.

Querying!

Guidance

A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.