# How to merge hhold\_train and indiv\_train datasets in r

**URL:** <https://community.drivendata.org/t/how-to-merge-hhold-train-and-indiv-train-datasets-in-r/1837>\
**Category:** Pover-T Tests: Predicting Poverty\
**Created:** [January 24, 2018, 8:26pm UTC](https://community.drivendata.org/t/how-to-merge-hhold-train-and-indiv-train-datasets-in-r/1837 "2018-01-24T20:26:28Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![peddollakishor](https://avatars.discourse-cdn.com/v4/letter/p/e8c25b/32.png) [@peddollakishor](https://community.drivendata.org/u/peddollakishor)\
**Post date:** [January 24, 2018, 8:26pm UTC](https://community.drivendata.org/t/how-to-merge-hhold-train-and-indiv-train-datasets-in-r/1837/1 "2018-01-24T20:26:28Z")

</div>

Hi All,

Anyone, please tell me how to merge hhold\_train and indiv\_train datasets in r.

Regards,  
Kishore

---

<div class="post-metadata">

**Author:** ![sagol](https://yyz2.discourse-cdn.com/flex028/user_avatar/community.drivendata.org/sagol/32/623_2.png) [@sagol](https://community.drivendata.org/u/sagol)\
**Post date:** [January 25, 2018, 2:13pm UTC](https://community.drivendata.org/t/how-to-merge-hhold-train-and-indiv-train-datasets-in-r/1837/2 "2018-01-25T14:13:54Z")

</div>

You can create new features from indiv\_train and add to hhold\_train.

---

<div class="post-metadata">

**Author:** ![pratikkgandhi](https://avatars.discourse-cdn.com/v4/letter/p/54ee81/32.png) [@pratikkgandhi](https://community.drivendata.org/u/pratikkgandhi)\
**Post date:** [January 26, 2018, 2:54am UTC](https://community.drivendata.org/t/how-to-merge-hhold-train-and-indiv-train-datasets-in-r/1837/3 "2018-01-26T02:54:15Z")

</div>

As Sagol mentioned, you can create features in hhold\_train and merge with indiv\_train dataset on ‘id’ column.

---

<div class="post-metadata">

**Author:** ![peddollakishor](https://avatars.discourse-cdn.com/v4/letter/p/e8c25b/32.png) [@peddollakishor](https://community.drivendata.org/u/peddollakishor)\
**Post date:** [January 28, 2018, 6:35pm UTC](https://community.drivendata.org/t/how-to-merge-hhold-train-and-indiv-train-datasets-in-r/1837/4 "2018-01-28T18:35:37Z")

</div>

@sagol thanks for reply.  
then i need to build models for both datasets right, my assumption is correct @sagol, please tell me if i am wrong.

---

<div class="post-metadata">

**Author:** ![sagol](https://yyz2.discourse-cdn.com/flex028/user_avatar/community.drivendata.org/sagol/32/623_2.png) [@sagol](https://community.drivendata.org/u/sagol)\
**Post date:** [January 28, 2018, 7:02pm UTC](https://community.drivendata.org/t/how-to-merge-hhold-train-and-indiv-train-datasets-in-r/1837/5 "2018-01-28T19:02:02Z")

</div>

When you add new features from indiv\_train to hhold\_train dataset you’ll get only one combined dataset for modelling.

---

<div class="post-metadata">

**Author:** ![peddollakishor](https://avatars.discourse-cdn.com/v4/letter/p/e8c25b/32.png) [@peddollakishor](https://community.drivendata.org/u/peddollakishor)\
**Post date:** [January 28, 2018, 7:28pm UTC](https://community.drivendata.org/t/how-to-merge-hhold-train-and-indiv-train-datasets-in-r/1837/6 "2018-01-28T19:28:34Z")

</div>

@sagol got it now, please have a look below r code and output image,  
library(reshape2)  
data\_w\<- dcast(a\_indiv\_train, id ~ HeUgMnzF,value.var=“iid”)

output:

 ![2018-01-29 00_55_33-F__drivendata - master - RStudio](https://canada1.discourse-cdn.com/flex028/uploads/drivendata1/original/1X/39d7107f63c2bbb54fc6915323ab5461e4f29205.jpg)

is this correct way to do, please correct me if i am do any mistake