dlakelan

using Distributions,Random,StatsPlots,LinearAlgebra


dist = MvNormal(zeros(20),I(20))
samp = rand(dist,200)

lps = [logpdf(dist,c) for c in eachcol(samp)]

histogram(lps; label="lp of 200 draws")

dlakelan

## from https://statmodeling.stat.columbia.edu/2024/03/07/relating-t-statistics-and-the-relative-width-of-confidence-intervals/#comment-2335932
# but fixed up due to blog damage

nsim = 10000

lengthOfCI = rep(NaN, nsim)
containsPM = rep(NaN, nsim)

for(i in 1:nsim){
    y = rnorm(10, 0, 1)

dlakelan

using HCubature, LinearAlgebra, StatsPlots, QuadGK, StaticArrays



# Define a function that describes the density 
# as a function of the radial distance from the center of the galaxy. This is a
# dimensionless quantity (mass/area) / (mass/area at zero distance)
# r is a dimensionless distance, radius as a fraction of the galactic scale radius 
function rho(x) 
    exp(-sqrt(x[1]^2+x[2]^2)-5.0*abs(x[3]))

dlakelan

using Distributions, Random, StatsPlots, Turing


# We have a pallet of 100 jugs of Orange Juice, each is a 2L jug.
# they are filled by a machine which may have a flaw. we would like to find
# the avg amount of OJ in a jug by taking a sample of 20
# this is a finite population, and the actual distribution of volumes is just given
# by 100 numbers. we'll say they are something like this:

Random.set_global_seed!(1234)

dlakelan

using Pkg
Pkg.activate(".")
Pkg.add(["Turing","Distributions","ApproxFun","QuadGK","StatsPlots","LinearAlgebra","DataFrames"])

using Turing, Distributions, ApproxFun, QuadGK, StatsPlots, LinearAlgebra, DataFrames

# a prior over functions that go up then come down... First let's see how ApproxFun works

Random.set_global_seed!(123456)
f = Fun(Chebyshev(Interval(0..100)),randn(12))

dlakelan

using Distributions, HypothesisTests, DataFrames, DataFramesMeta, StatsPlots

function dowork(effsizes,reps)
    res = typeof((effsize=1.0,meanval=1.0,pval=0.1))[]
    for s in effsizes
        for r in 1:reps
            sam = rand(Normal(s,1.0),15)
            tt = OneSampleTTest(sam)
            push!(res,(effsize=s,meanval=mean(sam),pval=pvalue(tt)))
        end

dlakelan

# julia code to show that the mean of 14 samples from a 50/50 mixture
# of normal and uniform has extremely close to "normal" distribution

using Distributions, StatsPlots

meanvals = [mean([rand(Normal(0,1),7); rand(Uniform(-1,1),7)]) for i in 1:100000]
histogram(meanvals)

pop = [rand(Normal(0,1),10000); rand(Uniform(-1,1),10000)];

dlakelan

# A simple stateful firewall with some packet tagging,
# based originally on nftables archlinux wiki
# https://wiki.archlinux.org/index.php/nftables

## this assumes eth0 is LAN and eth1 is WAN, modify as needed

flush ruleset

## change these