Rajan Sharma

• PhD Student at Macquarie University

Hi guys,

Instead of doing tedious job of writing up all conditions, I am trying a simple STATA loop such that a 'replace' command works if all variables satisfy an assigned criteria. My loop does work as such the replacement happens if any one of the variables fulfill a given condition. Any idea regarding this? My loop looks something like this:

gen xyz=.

forvalues i=0/10{

foreach v of varlist X1-X100{

replace xyz=1 if `v'[`i']>=5&`v'[`i']<=7&`v'<.

}

}

Thanks in advance.

I have this little (bizarre) problem. Having created a R markdown file and connecting it to scheduler via batch file, I want the output file to be named along with the date in which the batch runs..eg. xx.2016-03-03.html. 

The problem arises when attaching file in email system created using  mailR package. Manually, I could simply change the name of the file (for example: attach.files="C:/cc/xyz/xx.2016-03-16.html") which is not something I want. I want the above statement to happen through command and not manually.

Is there a way to solve this problem?

Regards,

Rajan

Hi all, I have created html slides from Rstudio presentation (.Rpres file) after combining R with Mysql. Now I want to automate the whole process such that .html output can be sent daily to desired emails. I think I can schedule the file through windows scheduler but I need some input with Rscript/batch file. Does somebody have experience of creating Rscript.exe (batch) file such that it captures .html output and email function to email that output?

Thank you in advance.

Regards

Hi everyone,

I have tried to collapse multiple rows per id in terms of the maximum values. I can do it using dplyr command:

data<-tbl_df(data) %>%

group_by(id) %>%

summarise_each(funs(max(., na.rm = TRUE)))

However, for a large data, it is little too slow. I tried the following using data.table but it produces wrong result:

data<-setDT(data)[, lapply(.SD, max,na.rm=TRUE), by=id]

I shall be grateful for suggestions.

Regards,

Rajan

I have large data tables in MySQL and is connected to R. I am using sqldf package in R to read tables from SQL and combining required tables to obtain the final table to carry out further analyses. But this is not very efficient as it occupies memory in R as well as it takes long time to extract big tables.

Is there a package or something in R that will help me work by directly referencing to tables in MySQL without having to read them first in R?

Thanks!

Hi friends, I have to find possible matches in a large database for a small number of cases (on the basis of many variables).  There are many missing values in the variables which make it unfit for matching (I am doing this using 'matchit' package in R). That's why I did imputation first and did matching after that but I have doubts surrounding this. My question is: Is imputation the best way to address missing values before matching? Is there a matching technique that ignores the missing values and matches irrespective of whether there are missing values or not? (R and/or STATA)

Hi friends,

I am working in a reproducible presentation in R where I've to show some numbers in the form of a flow chart. I tried some packages like DiagrammeR but they give error messages when I put in the R chunk.

Is there any other way?