Statistical Concepts: Difference between revisions

From NovaOrdis Knowledge Base
Jump to navigation Jump to search
No edit summary
 
(25 intermediate revisions by the same user not shown)
Line 5: Line 5:
=Concepts=
=Concepts=


* mean
* [[Descriptive_Statistics#Overview|Descriptive Statistics]]
 
* Mean/Median/Mode. Difference between mean and average.
* Unique Median
* standard deviation
* standard deviation
* linear regression
* <span id='Regression'></span>[[Regression|Regression]] and [[Regression#Linear_Regression|Linear Regression]].
* correlation
** Dependent variable (criterion)
** Independent variable (predictor)
 
* [[Classification]]
 
* [[Bayes Rule]]
* [[Bayes Rule]]
* [[Percentile]]
* Scatter Plot
* Linearity (linear exact or not exact)
* Positive and negative linear relationship.
* Outlier
* Deviation
* Noise - deviation from a linear graph.
* Monotonicity.
* Bar Charts. Applies to 2D data.
* Global trends.
* Historgram. A bar chart where the vertical axis is a frequency count, as a function of the range. Applies to 1D data.
* Frequency Count
* Pie charts - represent relative outcomes.
* Unrelated data
* Simpson's paradox
* Be skeptical and really understand how to turn raw data into conclusions.
* Probability - the opposite of statistics.
* P() notation
* Truth table
* Probability of a composite event (independence)
* Dependence
* Conditional probability
* Conditional probability notation - important for Bayes Rule
* Total probability
* [[Bayes Rule]]
** Prior probability
** Unreliable measurement (Sensitivity/Specificity)
** Joint probabilty
** Posterior probabilty
* Probability Distribution
* Continous Probability Distribution.
* Density of probability
* Estimators
* Laplacian estimator
* Empirical (observational) frequency
* Maximum likelihood estimator
* Dirichelet data
* Laplacian Estimator
* Mode, bimodal, multimodal
* Variance
* Standard Deviation
* Standard Score
* [[Time Series]]
==Correlation and Causation==
* Correlation vs. Causation
* Variables
* Definition of correlation (Is correlation injectivity?)
* Confounding variable.
* https://en.wikipedia.org/wiki/Correlation_does_not_imply_causation
* Even if there is no causation, correlation can be used in prediction.
=TODO=
* Relocate Continous Functions
* Granger causality test https://en.wikipedia.org/wiki/Granger_causality

Latest revision as of 22:28, 14 May 2024

Internal

Concepts

  • Mean/Median/Mode. Difference between mean and average.
  • Unique Median
  • standard deviation
  • Regression and Linear Regression.
    • Dependent variable (criterion)
    • Independent variable (predictor)
  • Bayes Rule
  • Percentile
  • Scatter Plot
  • Linearity (linear exact or not exact)
  • Positive and negative linear relationship.
  • Outlier
  • Deviation
  • Noise - deviation from a linear graph.
  • Monotonicity.
  • Bar Charts. Applies to 2D data.
  • Global trends.
  • Historgram. A bar chart where the vertical axis is a frequency count, as a function of the range. Applies to 1D data.
  • Frequency Count
  • Pie charts - represent relative outcomes.
  • Unrelated data
  • Simpson's paradox
  • Be skeptical and really understand how to turn raw data into conclusions.
  • Probability - the opposite of statistics.
  • P() notation
  • Truth table
  • Probability of a composite event (independence)
  • Dependence
  • Conditional probability
  • Conditional probability notation - important for Bayes Rule
  • Total probability
  • Bayes Rule
    • Prior probability
    • Unreliable measurement (Sensitivity/Specificity)
    • Joint probabilty
    • Posterior probabilty
  • Probability Distribution
  • Continous Probability Distribution.
  • Density of probability
  • Estimators
  • Laplacian estimator
  • Empirical (observational) frequency
  • Maximum likelihood estimator
  • Dirichelet data
  • Laplacian Estimator
  • Mode, bimodal, multimodal
  • Variance
  • Standard Deviation
  • Standard Score


Correlation and Causation

TODO