links for research - didn't end up getting to them today

the other day i learnt about the electron application wrapper which could be used to wrap full-stack applications and turn them into desktop executables - this is something i feel would be useful in my future personal projects

however, today i was more focused on other things on my checklist so i'll link the two handy links i found so i can come back to them:

• using electron to make a web app into a desktop executable
• electron docs

freecodecamp: react and redux

at 12.48 i finished the react and redux course.

however, as someone who has never used react nor redux in a project before, i was quite confused trying to follow the tutorial and almost all the code i ended up uploading to my personal freecodecamp learning github repo were the solutions to each lesson of the course

datacamp: sampling in python - completed!!

finished the day with datacamp, ending at 20.22. here are my notes for what i covered:

chapter 3: sampling distributions

• relative error of point estimates
  • sample size is the number of rows / observations in the sample
  • relative error is the difference between the population mean (population parameter) and sample mean (point estimate)
• standard errors and the central limit theorem
  • consequences of the central limit theorem - averages of independent samples have approximately normal distributions
  • as the sample size increases, distribution of averages gets closer to being normally distributed, and width of the sampling distribution gets narrower

chapter 4: bootstrap distributions

• introduction to bootstrapping
  • bootstrapping - the opposite of sampling from a population; it’s building up a theoretical population from the sample
  • bootstrapping is used to develop an understanding of sampling variability using a single sample
  • bootstrapping steps
    • make resample of same size as original sample
    • calculate statistic of interest for bootstrap sample
    • repeat steps 1 and 2 many times
    • resulting statistics are bootstrap statistics that form a bootstrap distribution
• comparing sampling and bootstrap distributions
  • interpreting the bootstrap distribution mean
    • usually close to sample mean
    • may not be a good estimate of the population mean
    • repeat steps 1 and 2 many times
    • cannot correct biases from sampling
  • standard error is the standard deviation of the statistic of interest
  • estimated standard error is the stand deviation of the bootstrap distribution for a sample statistic
• confidence intervals
  • the values within one standard deviation of the mean