Statistics

Pages in this section

  • Accuracy
    Last edited: 2026-02-05

    Accuracy

    Suppose we some model $\hat{f}$ predicting $f$. For some test data $T$ we define

  • Expected value
    Last edited: 2026-02-05

    Expected value

    For some random variable $X$ with domain $D$ (with a sense of addition and multiplication by $\mathbb{R}$) and probability density function $f: D \rightarrow [0,1] \subset \mathbb{R}$ the expected value is:

  • Introduction to statistical learning with Applications in Python
    Last edited: 2024-01-11

    This is a book by multiple authors about Machine Learning I was reading with a bunch of my friends. You can find it in my repo or for free on their website .

  • Maximum a posteriori probability estimate (MAP)
    Last edited: 2026-02-05

    Maximum a posteriori probability estimate (MAP)

    Suppose we have a hypothesis space $H$ and we want to pick the best hypothesis given some data $D$. Further more suppose we have prior belief about the likelihood of each hypothesis represented by a probability distribution over $H$. The maximum a posteriori probability estimate is

  • Maximum likelihood estimation (MLE)
    Last edited: 2026-02-05

    Maximum likelihood estimation (MLE)

    Suppose we have a hypothesis space $H$ and we want to pick the best hypothesis given some data $D$. The maximum likelihood estimation is

  • Modelling framework
    Last edited: 2026-02-05

    Modelling framwork

    Suppose we have some random variable we want to predict $Y$ (over a space $B$) and some set of features or predictors in $A$ to make predictions of $Y$ this are sampled from a random variable $X$. You assume there is some relationship between $Y$ and $X$ given by

  • Normal distribution
    Last edited: 2026-02-05

    Normal distribution

    The normal distribution is a continuous probability distribution $f: \mathbb{R} \rightarrow (0,1) \subset \mathbb{R}$ specified by the mean $\mu$ and standard deviation $\sigma$ given by probability density function

  • p-value
    Last edited: 2024-01-10

    Assuming the Null hypothese, what is the probability of seeing a specific result? If data is on the “not expected” region, we reject the null hypothesis.

  • Recall
    Last edited: 2026-02-05

    Recall

    For some binary classification problem where we are using $\hat{f}: A \rightarrow \{1, -1\}$ to predict $f: A \rightarrow \{1, -1\}$ for some testing data $T$ we define

  • Result types
    Last edited: 2026-02-05

    Result types

    In a binary classification there you model has 4 outcomes for a prediction. results_explanation This is explained by two binary decisions. Did it guess it to be positive or negative and was it correct. If it was correct it is True …. if not it was False … . If it guessed it to be possitive it is … positive if not it was …. negative. This combined provides true positive, true negative, false positive, and false negative.