diff --git a/Kandi.pdf b/Kandi.pdf index 1facc56b54cca32436a80631a974e6dc84ad3841..661d9e394f5e6a459f91742cbe7edf1784c572f9 100644 Binary files a/Kandi.pdf and b/Kandi.pdf differ diff --git a/Kandi.synctex.gz b/Kandi.synctex.gz index dc722e5833bb2d2381319c53779cb24561b2a1e9..6eb2f692f28996592bcd0045aa66e9742749b315 100644 Binary files a/Kandi.synctex.gz and b/Kandi.synctex.gz differ diff --git a/Kandi.tex b/Kandi.tex index 4950e9113424cbf2f7e60758f756f53416bcbbb5..f575705e23b4cadf23166aeab6d16078b7640987 100644 --- a/Kandi.tex +++ b/Kandi.tex @@ -75,7 +75,7 @@ \addtolength{\voffset}{0.45cm} \addtolength{\textheight}{-0.9cm} -\title{Kandidaatintutkielma\\ {\Large Kausaalipäättely valikoitumisharha korjaamisessa}} % Parempi otsikko +\title{Kandidaatintutkielma\\ {\Large Kausaalipäättely valikoitumisharhan korjaamisessa}} % Parempi otsikko \author{Riku Laine\\ Valtiotieteellinen tiedekunta \\ Helsingin yliopisto} \date{\today} diff --git a/paper/sl.pdf b/paper/sl.pdf new file mode 100644 index 0000000000000000000000000000000000000000..64658b57f1acae11a6fb9a8196df9a80466eb569 Binary files /dev/null and b/paper/sl.pdf differ diff --git a/paper/sl.synctex.gz b/paper/sl.synctex.gz new file mode 100644 index 0000000000000000000000000000000000000000..7f7d7ad46b7b38cfc3fd6d39fb8d7ce93b5aa041 Binary files /dev/null and b/paper/sl.synctex.gz differ diff --git a/paper/sl.tex b/paper/sl.tex index faad3ccfcf85264673cb8f8569f64512a7a698e9..03c1cbc1bd106874c3c01a37540034d499f9ce56 100755 --- a/paper/sl.tex +++ b/paper/sl.tex @@ -98,26 +98,31 @@ The decision is based on the following variables. First, the features \features Secondly, the leniency of the judge, expressed as a variable \leniency. Specifically, we assume that every judge evaluates a given candidate according to the probability \[ -\prob{\outcome = 1 | \features = \featuresValue, \doop{\decision = 1}} +\prob{\outcome = 0 | \features = \featuresValue, \doop{\decision = 1}} \] -that the candidate will violate bail conditions (\outcome = 1) if they were granted bail. +that the candidate will violate bail conditions (\outcome = 0) if they were granted bail. We write \outcome = 0 to refer to the case when the defendant does not violate bail, whether bail is granted or not. The \doop{condition} expression signifies that, in evaluating the probability, we consider the event where the condition (here, it is the condition $\decision = 1$) is imposed to the data-generation process (and therefore alters the generative model). In addition, we assume that every judge would assign the same value to the above probability, given by a function \score{\featuresValue}. \[ -\score{\featuresValue} = \prob{\outcome = 1 | \features = \featuresValue, \doop{\decision = 1}} +\score{\featuresValue} = \prob{\outcome = 0 | \features = \featuresValue, \doop{\decision = 1}} \] The assumption that, essentially, all judges have the same model for the probability that a defendant would violate bail is not far-fetched for the purposes of our analysis, particularly taking into account that \score{\featuresValue} can be learned from the observed data \[ -\prob{\outcome = 1 | \features = \featuresValue, \doop{\decision = 1}} = \prob{\outcome = 1 | \features = \featuresValue, \decision = 1} +\prob{\outcome = 0 | \features = \featuresValue, \doop{\decision = 1}} = \prob{\outcome = 0 | \features = \featuresValue, \decision = 1} \] and that data are publicly accessible, allowing us to assume that all judges have access to the same information. Where judges {\it do differ} is at the level of their leniency \leniency. Following the above assumptions, a judge with leniency \leniency = \leniencyValue grants bail to the defendants for which $F(\featuresValue) < r$, where $F$ is the cumulative distribution. \begin{equation} - F(\featuresValue_0) = \int { \indicator{\prob{\outcome = 1| \decision = 1, \features = \featuresValue} > \prob{\outcome = 1| \decision = 1, \features = \featuresValue_0}} d\prob{\featuresValue} } + F(\featuresValue_0) = \int { \indicator{\prob{\outcome = 0| \decision = 1, \features = \featuresValue} > \prob{\outcome = 0| \decision = 1, \features = \featuresValue_0}} d\prob{\featuresValue} } \end{equation} +which should be equal to + +\begin{equation} + F(\featuresValue_0) = \int {\prob{\featuresValue} \indicator{\prob{\outcome = 0| \decision = 1, \features = \featuresValue} > \prob{\outcome = 0| \decision = 1, \features = \featuresValue_0}} d\featuresValue} +\end{equation} The bail-or-jail scenario is just one example of settings that involve a decision $\decision \in\{0,1\}$ that is based on individual features \features and leniency (acceptance rate) \leniency -- and where a behavior of interest \outcome is observed only for the cases where \decision = 1. The diagram of the causal model is shown in Figure~\ref{fig:causalmodel}. Our results are applicable to other scenarios with same causal model. @@ -138,23 +143,23 @@ Performance is measured {\it for a given leniency level} as the rate at which ba In other words, performance is measured as the probability that a decision lead to undesired outcome. \section{Analysis} -We wish to calculate the probability of undesired outcome (\outcome = 1) at a fixed leniency level. +We wish to calculate the probability of undesired outcome (\outcome = 0) at a fixed leniency level. \begin{align*} -& \prob{\outcome = 1 | \doop{\leniency = \leniencyValue}} = \nonumber \\ -& = \sum_\decisionValue \prob{\outcome = 1, \decision = \decisionValue | \doop{\leniency = \leniencyValue}} \nonumber \\ -& = \prob{\outcome = 1, \decision = 0 | \doop{\leniency = \leniencyValue}} + \prob{\outcome = 1, \decision = 1 | \doop{\leniency = \leniencyValue}} \nonumber \\ -& = 0 + \prob{\outcome = 1, \decision = 1 | \doop{\leniency = \leniencyValue}} \nonumber \\ -& = \prob{\outcome = 1, \decision = 1 | \doop{\leniency = \leniencyValue}} \nonumber \\ -& = \sum_\featuresValue \prob{\outcome = 1, \decision = 1, \features = \featuresValue | \doop{\leniency = \leniencyValue}} \nonumber \\ -& = \sum_\featuresValue \prob{\outcome = 1, \decision = 1 | \doop{\leniency = \leniencyValue}, \features = \featuresValue} \prob{\features = \featuresValue | \doop{\leniency = \leniencyValue}} \nonumber \\ -& = \sum_\featuresValue \prob{\outcome = 1, \decision = 1 | \doop{\leniency = \leniencyValue}, \features = \featuresValue} \prob{\features = \featuresValue} \nonumber \\ -& = \sum_\featuresValue \prob{\outcome = 1 | \decision = 1, \doop{\leniency = \leniencyValue}, \features = \featuresValue} \prob{\decision = 1 | \doop{\leniency = \leniencyValue}, \features = \featuresValue} \prob{\features = \featuresValue} \nonumber \\ -& = \sum_\featuresValue \prob{\outcome = 1 | \decision = 1, \features = \featuresValue} \prob{\decision = 1 | \leniency = \leniencyValue, \features = \featuresValue} \prob{\features = \featuresValue} +& \prob{\outcome = 0 | \doop{\leniency = \leniencyValue}} = \nonumber \\ +& = \sum_\decisionValue \prob{\outcome = 0, \decision = \decisionValue | \doop{\leniency = \leniencyValue}} \nonumber \\ +& = \prob{\outcome = 0, \decision = 0 | \doop{\leniency = \leniencyValue}} + \prob{\outcome = 0, \decision = 1 | \doop{\leniency = \leniencyValue}} \nonumber \\ +& = 0 + \prob{\outcome = 0, \decision = 1 | \doop{\leniency = \leniencyValue}} \nonumber \\ +& = \prob{\outcome = 0, \decision = 1 | \doop{\leniency = \leniencyValue}} \nonumber \\ +& = \sum_\featuresValue \prob{\outcome = 0, \decision = 1, \features = \featuresValue | \doop{\leniency = \leniencyValue}} \nonumber \\ +& = \sum_\featuresValue \prob{\outcome = 0, \decision = 1 | \doop{\leniency = \leniencyValue}, \features = \featuresValue} \prob{\features = \featuresValue | \doop{\leniency = \leniencyValue}} \nonumber \\ +& = \sum_\featuresValue \prob{\outcome = 0, \decision = 1 | \doop{\leniency = \leniencyValue}, \features = \featuresValue} \prob{\features = \featuresValue} \nonumber \\ +& = \sum_\featuresValue \prob{\outcome = 0 | \decision = 1, \doop{\leniency = \leniencyValue}, \features = \featuresValue} \prob{\decision = 1 | \doop{\leniency = \leniencyValue}, \features = \featuresValue} \prob{\features = \featuresValue} \nonumber \\ +& = \sum_\featuresValue \prob{\outcome = 0 | \decision = 1, \features = \featuresValue} \prob{\decision = 1 | \leniency = \leniencyValue, \features = \featuresValue} \prob{\features = \featuresValue} \end{align*} Expanding the above derivation for model \score{\featuresValue} learned from the data \[ -\score{\featuresValue} = \prob{\outcome = 1 | \features = \featuresValue, \decision = 1}, +\score{\featuresValue} = \prob{\outcome = 0 | \features = \featuresValue, \decision = 1}, \] the {\it generalized performance} \generalPerformance of that model is given by the following formula. \begin{equation}