#quick statistics question

1 messages · Page 1 of 1 (latest)

lapis stirrup
#

For context, I take this course about using MatLab, which is basically using code to solve basic mathematics, And I've been assigned a project to solve any problem I could think of. Because I play games and like seeing numbers of what I do, I wanted to know how much in-game currency I could get from missions in a month.

So I've been collecting data for the past month to build my estimation on, considering that:
I get 80-130 from daily quests and I get one daily quest a day, 80 is way more common than 130, and rarity increases with the amount I'll get
I get 0-80 from mission alerts a day, yet its quite rare to get more than 50 a day, and its very common to just get 0

my question is how can I use this excel sheet of the past months data to get an estimate of how much I'll get monthly, and I want to take into account an error chance too?

slender oceanBOT
#
  1. Ask your question and show the work you've done so far. If you've posted a screenshot of a question, specify which part you need help with.
  2. Wait patiently for a helper to come along.
  3. Once someone helps you, say thank you and close the thread with:
    +close
  4. Feel free to nominate the person for helper of the week in #helper-nominations
  5. Do not ping the mods, unless someone is breaking the rules.
  6. If you're happy with the help you got here, and the server overall, you can contribute financially as well:
old ivy
#

eyeball it, call it like 90, you need to measure this income thousands of times to conclude anything reliably

lapis stirrup
#

so would a mean value function work or do i need more complicated maths?

#

ive looked into websites that keep track of this stuff on a mission alert level, getting monthly data, and its within the range of 800-900 a month too

old ivy
#

then you can quote that data

lapis stirrup
#

but i have to do the math in the actual code too

old ivy
#

that's just syntax

lapis stirrup
#

makes sense

old ivy
#

nobody will

lapis stirrup
#

yeah, plus its a game so its subject to change at any given moment

old ivy
#

and that, regardless, if you already have data available, then use it to do your computations

lapis stirrup
pseudo granite
#

I don't mean to piss on your work, but I'll explain my reasoning

#

One can model your daily quest rewards as following a distribution $P_X$ and your mission alert rewards as following a distribution $P_Y$

zenith topazBOT
pseudo granite
#

And throughout a period of $n$ days, you basically sample $n$ independent $X_1, ..., X_n \sim_{i.i.d.} P_X$ and $Y_1, ..., Y_n \sim_{i.i.d.} P_Y$

zenith topazBOT
pseudo granite
#

Now see that your target variable $T = \sum_{i=1}^{n} X_i + \sum_{i=1}^{n} Y_i$ is a sum of independent random variables

zenith topazBOT
pseudo granite
#

So it is clear that $E[T] = n(E[X_1] + E[Y_1])$

zenith topazBOT
pseudo granite
#

however, $V(T) = n V(X_1) + n V(Y_1)$ as well

zenith topazBOT