Hello, an idea!

Hey everyone,

I’m a junior in college, majoring in math and statistics. Today I started something I’ve been planning for a long time, I just finally got around to doing it. I found this forum online and I’m really excited to find a community like this! I’m gathering data on everything about myself in each day (mood, food intake, physical activity, work/educational activity). The mood, food intake, and physical activity are all covered so far using a variety of iphone apps, the work/educational activity is something I haven’t developed a complete system for yet. I’m thinking of something along the lines of estimating the time an activity will take, timing myself on activities, recording type of activity, time of day of the activity, etc… to get an estimate of my productivity levels, as well as a sort of estimation system which would help me predict how long something would take (this would be ideal for repetitive tasks like homework).

The end goal of gathering this data is to better understand how certain habits affect my productivity and mood levels, and hopefully eventually try and optimize them.

I started making a web app a couple weeks ago to measure productivity, but I didn’t finish it (a lot of other things needed to be done). I’m making it using R shiny. I’m sure a lot of you are familiar with R, and it’s the language I’m most proficient in. R shiny basically allows you to make web apps to visualize and perform analysis on data. Examples: http://shiny.rstudio.com/gallery/

I apologize if this topic has been brought up before, I haven’t had a chance to read through the entire forum. R shiny is ideal for this sort of thing in my opinion, since all of the backend data analysis and visualization tools are built into R. I’ve made several R shiny apps so far, and have a good working knowledge of how it all works.

Another idea I had is to use interval data instead of exact numbers. (See: http://en.wikipedia.org/wiki/Interval_arithmetic)

I’m doing some research with a statistician in this field, and I’ve been convinced that interval analysis is a very powerful technique when dealing with uncertain data (like mood estimates or nutritional facts).

Are you doing some kind of multiple regression analysis, or are you doing a series of experiments where you change a single variable at a time?