Is it ever recommended to use mean/multiple imputation when using tree-based predictive models?Orthogonal...

Why is a white electrical wire connected to 2 black wires?

Is it insecure to send a password in a `curl` command?

Does .bashrc contain syntax errors?

How to pronounce "I ♥ Huckabees"?

Do the common programs (for example: "ls", "cat") in Linux and BSD come from the same source code?

Why one should not leave fingerprints on bulbs and plugs?

How to make healing in an exploration game interesting

How do I hide Chekhov's Gun?

Are ETF trackers fundamentally better than individual stocks?

Most cost effective thermostat setting: consistent temperature vs. lowest temperature possible

Are all passive ability checks floors for active ability checks?

Are Roman Catholic priests ever addressed as pastor

Relationship between sampajanna definitions in SN 47.2 and SN 47.35

What are substitutions for coconut in curry?

What is a ^ b and (a & b) << 1?

Adventure Game (text based) in C++

How are passwords stolen from companies if they only store hashes?

Four married couples attend a party. Each person shakes hands with every other person, except their own spouse, exactly once. How many handshakes?

Examples of transfinite towers

How to write cleanly even if my character uses expletive language?

Explaining pyrokinesis powers

Violin - Can double stops be played when the strings are not next to each other?

Bacteria contamination inside a thermos bottle

Custom alignment for GeoMarkers



Is it ever recommended to use mean/multiple imputation when using tree-based predictive models?


Orthogonal sets of variables in multiple imputation --> separate imputation models?Multiple Imputation Using Different Data Setsusing cluster information in multiple imputationMultiple Imputation for Spatial Modelsmultiple imputation models containing categorical variablesWhen to use multiple imputation chained equations vs regression to impute data?Multiple imputation when explained variance of imputation model is lowPredictive Mean Matching as Single Imputation?How to apply a model built using Multiple Imputation to predict on dataset with missing data?How NULLs in numerical variables are treated in tree-based models?













3












$begingroup$


Everytime that I am making some predictive model and I have missing data I impute categorical variables with something like "UNKNOWN" and numerical variables with some absurd number that will never be seen in practice (even if the variable is unbounded I can take the exponent of the variable and make the unknown values negative).



The main advantage is that the model knows that the variable is missing, which is not the case for say mean imputation. I can see that this could be disastrous in linear models or neural networks but in tree-based models this is handled really smoothly.



I know that there is a great deal of literature on missing data imputation, but when and why would I ever use these methods when missing data for predictive (tree-based) models?










share|cite|improve this question









$endgroup$












  • $begingroup$
    Imputing a large number for numeric data could be very bad for tree based models. Think of it this way, if your split is for example on income and the split is at say 100k, now everyone that was missing is going to be in the split with the high income earners
    $endgroup$
    – astel
    1 hour ago










  • $begingroup$
    The model will be fitted with that imputed values as well - so if they are significantly different than people with true high income the tree should make a split with true high and fake high (missing) income. If variability is low inside the tree node then there is not much to worry.
    $endgroup$
    – gsmafra
    1 hour ago
















3












$begingroup$


Everytime that I am making some predictive model and I have missing data I impute categorical variables with something like "UNKNOWN" and numerical variables with some absurd number that will never be seen in practice (even if the variable is unbounded I can take the exponent of the variable and make the unknown values negative).



The main advantage is that the model knows that the variable is missing, which is not the case for say mean imputation. I can see that this could be disastrous in linear models or neural networks but in tree-based models this is handled really smoothly.



I know that there is a great deal of literature on missing data imputation, but when and why would I ever use these methods when missing data for predictive (tree-based) models?










share|cite|improve this question









$endgroup$












  • $begingroup$
    Imputing a large number for numeric data could be very bad for tree based models. Think of it this way, if your split is for example on income and the split is at say 100k, now everyone that was missing is going to be in the split with the high income earners
    $endgroup$
    – astel
    1 hour ago










  • $begingroup$
    The model will be fitted with that imputed values as well - so if they are significantly different than people with true high income the tree should make a split with true high and fake high (missing) income. If variability is low inside the tree node then there is not much to worry.
    $endgroup$
    – gsmafra
    1 hour ago














3












3








3


1



$begingroup$


Everytime that I am making some predictive model and I have missing data I impute categorical variables with something like "UNKNOWN" and numerical variables with some absurd number that will never be seen in practice (even if the variable is unbounded I can take the exponent of the variable and make the unknown values negative).



The main advantage is that the model knows that the variable is missing, which is not the case for say mean imputation. I can see that this could be disastrous in linear models or neural networks but in tree-based models this is handled really smoothly.



I know that there is a great deal of literature on missing data imputation, but when and why would I ever use these methods when missing data for predictive (tree-based) models?










share|cite|improve this question









$endgroup$




Everytime that I am making some predictive model and I have missing data I impute categorical variables with something like "UNKNOWN" and numerical variables with some absurd number that will never be seen in practice (even if the variable is unbounded I can take the exponent of the variable and make the unknown values negative).



The main advantage is that the model knows that the variable is missing, which is not the case for say mean imputation. I can see that this could be disastrous in linear models or neural networks but in tree-based models this is handled really smoothly.



I know that there is a great deal of literature on missing data imputation, but when and why would I ever use these methods when missing data for predictive (tree-based) models?







missing-data cart boosting data-imputation multiple-imputation






share|cite|improve this question













share|cite|improve this question











share|cite|improve this question




share|cite|improve this question










asked 1 hour ago









gsmafragsmafra

16518




16518












  • $begingroup$
    Imputing a large number for numeric data could be very bad for tree based models. Think of it this way, if your split is for example on income and the split is at say 100k, now everyone that was missing is going to be in the split with the high income earners
    $endgroup$
    – astel
    1 hour ago










  • $begingroup$
    The model will be fitted with that imputed values as well - so if they are significantly different than people with true high income the tree should make a split with true high and fake high (missing) income. If variability is low inside the tree node then there is not much to worry.
    $endgroup$
    – gsmafra
    1 hour ago


















  • $begingroup$
    Imputing a large number for numeric data could be very bad for tree based models. Think of it this way, if your split is for example on income and the split is at say 100k, now everyone that was missing is going to be in the split with the high income earners
    $endgroup$
    – astel
    1 hour ago










  • $begingroup$
    The model will be fitted with that imputed values as well - so if they are significantly different than people with true high income the tree should make a split with true high and fake high (missing) income. If variability is low inside the tree node then there is not much to worry.
    $endgroup$
    – gsmafra
    1 hour ago
















$begingroup$
Imputing a large number for numeric data could be very bad for tree based models. Think of it this way, if your split is for example on income and the split is at say 100k, now everyone that was missing is going to be in the split with the high income earners
$endgroup$
– astel
1 hour ago




$begingroup$
Imputing a large number for numeric data could be very bad for tree based models. Think of it this way, if your split is for example on income and the split is at say 100k, now everyone that was missing is going to be in the split with the high income earners
$endgroup$
– astel
1 hour ago












$begingroup$
The model will be fitted with that imputed values as well - so if they are significantly different than people with true high income the tree should make a split with true high and fake high (missing) income. If variability is low inside the tree node then there is not much to worry.
$endgroup$
– gsmafra
1 hour ago




$begingroup$
The model will be fitted with that imputed values as well - so if they are significantly different than people with true high income the tree should make a split with true high and fake high (missing) income. If variability is low inside the tree node then there is not much to worry.
$endgroup$
– gsmafra
1 hour ago










1 Answer
1






active

oldest

votes


















2












$begingroup$

One reason you may not want to use "insert impossible value" methods is that means that your predictive model works conditional on the distribution of the data missingness remaining unchanged. Thus, if after building your tree model, it is realized that we can start using certain features more often, we can no longer use the model that was built using the "impute impossible value" method without retraining the model.



In fact, this problem is even further compounded if the rates of missingness changes during the data collection process itself. Then, even immediately after building the model, it is already "out of date", as the current rates of missingness will be different than the rates of missingness during when the data was collected.



To illustrate the issue, let's suppose a bank is building a database to help predict if clients will default on a loan. Early in the data collection process, loan officers have the option to conduct a background investigation, but they almost never do for clients they deem as trustworthy. Thus, for the especially trustworthy customers, the background check variable is almost always missing. If you use the "impute impossible value" method, having a possible value for background checks indicates high risk.



If background check rates don't change at all, then this "impute impossible value" method will likely still provide valid predictions. However, let's suppose the bank realizes that background checks are really helpful for assessing risk, so they change their policy to include background checks for everyone. Then, everyone will have a possible value for background checks and using the "impute impossible value" method, everyone will be flagged as "high risk".



Cross validation will not catch this issue, as the missingness distribution will be the same between the training and testing sets. So even though the "impute impossible value" method may lead to pretty results during cross-validation, this will lead to poor predictions upon deployment!



Note that you will essentially need to throw away all your data everytime your data collection policy changes! Alternatively, if you can correctly impute the missing values and their uncertainty, you can now use the data that was collected under the old policy.






share|cite|improve this answer











$endgroup$













  • $begingroup$
    That's a good point, imputation could be more robust on changes in the way data is missing. I will take your statement on throwing away past data as an exaggeration though - including a time variable and retraining the model should be enough make it usable again.
    $endgroup$
    – gsmafra
    32 mins ago










  • $begingroup$
    @gsmafra: In general, I don't think adding a time variable will fix the problem. For example, in a random forest, the time variable will only be included in 1/3 of the trees, so it won't even be included in the majority of the decision trees in your random forest.
    $endgroup$
    – Cliff AB
    24 mins ago










  • $begingroup$
    To be clear, I don't think you should throw out your data...but I'd only advise doing "impossible value imputation" on variables you don't think will be very predictive to start with or you're fairly certain that the missingness distribution is fairly stable.
    $endgroup$
    – Cliff AB
    23 mins ago













Your Answer





StackExchange.ifUsing("editor", function () {
return StackExchange.using("mathjaxEditing", function () {
StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix) {
StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
});
});
}, "mathjax-editing");

StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "65"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);

StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});

function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: false,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: null,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});


}
});














draft saved

draft discarded


















StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstats.stackexchange.com%2fquestions%2f397942%2fis-it-ever-recommended-to-use-mean-multiple-imputation-when-using-tree-based-pre%23new-answer', 'question_page');
}
);

Post as a guest















Required, but never shown

























1 Answer
1






active

oldest

votes








1 Answer
1






active

oldest

votes









active

oldest

votes






active

oldest

votes









2












$begingroup$

One reason you may not want to use "insert impossible value" methods is that means that your predictive model works conditional on the distribution of the data missingness remaining unchanged. Thus, if after building your tree model, it is realized that we can start using certain features more often, we can no longer use the model that was built using the "impute impossible value" method without retraining the model.



In fact, this problem is even further compounded if the rates of missingness changes during the data collection process itself. Then, even immediately after building the model, it is already "out of date", as the current rates of missingness will be different than the rates of missingness during when the data was collected.



To illustrate the issue, let's suppose a bank is building a database to help predict if clients will default on a loan. Early in the data collection process, loan officers have the option to conduct a background investigation, but they almost never do for clients they deem as trustworthy. Thus, for the especially trustworthy customers, the background check variable is almost always missing. If you use the "impute impossible value" method, having a possible value for background checks indicates high risk.



If background check rates don't change at all, then this "impute impossible value" method will likely still provide valid predictions. However, let's suppose the bank realizes that background checks are really helpful for assessing risk, so they change their policy to include background checks for everyone. Then, everyone will have a possible value for background checks and using the "impute impossible value" method, everyone will be flagged as "high risk".



Cross validation will not catch this issue, as the missingness distribution will be the same between the training and testing sets. So even though the "impute impossible value" method may lead to pretty results during cross-validation, this will lead to poor predictions upon deployment!



Note that you will essentially need to throw away all your data everytime your data collection policy changes! Alternatively, if you can correctly impute the missing values and their uncertainty, you can now use the data that was collected under the old policy.






share|cite|improve this answer











$endgroup$













  • $begingroup$
    That's a good point, imputation could be more robust on changes in the way data is missing. I will take your statement on throwing away past data as an exaggeration though - including a time variable and retraining the model should be enough make it usable again.
    $endgroup$
    – gsmafra
    32 mins ago










  • $begingroup$
    @gsmafra: In general, I don't think adding a time variable will fix the problem. For example, in a random forest, the time variable will only be included in 1/3 of the trees, so it won't even be included in the majority of the decision trees in your random forest.
    $endgroup$
    – Cliff AB
    24 mins ago










  • $begingroup$
    To be clear, I don't think you should throw out your data...but I'd only advise doing "impossible value imputation" on variables you don't think will be very predictive to start with or you're fairly certain that the missingness distribution is fairly stable.
    $endgroup$
    – Cliff AB
    23 mins ago


















2












$begingroup$

One reason you may not want to use "insert impossible value" methods is that means that your predictive model works conditional on the distribution of the data missingness remaining unchanged. Thus, if after building your tree model, it is realized that we can start using certain features more often, we can no longer use the model that was built using the "impute impossible value" method without retraining the model.



In fact, this problem is even further compounded if the rates of missingness changes during the data collection process itself. Then, even immediately after building the model, it is already "out of date", as the current rates of missingness will be different than the rates of missingness during when the data was collected.



To illustrate the issue, let's suppose a bank is building a database to help predict if clients will default on a loan. Early in the data collection process, loan officers have the option to conduct a background investigation, but they almost never do for clients they deem as trustworthy. Thus, for the especially trustworthy customers, the background check variable is almost always missing. If you use the "impute impossible value" method, having a possible value for background checks indicates high risk.



If background check rates don't change at all, then this "impute impossible value" method will likely still provide valid predictions. However, let's suppose the bank realizes that background checks are really helpful for assessing risk, so they change their policy to include background checks for everyone. Then, everyone will have a possible value for background checks and using the "impute impossible value" method, everyone will be flagged as "high risk".



Cross validation will not catch this issue, as the missingness distribution will be the same between the training and testing sets. So even though the "impute impossible value" method may lead to pretty results during cross-validation, this will lead to poor predictions upon deployment!



Note that you will essentially need to throw away all your data everytime your data collection policy changes! Alternatively, if you can correctly impute the missing values and their uncertainty, you can now use the data that was collected under the old policy.






share|cite|improve this answer











$endgroup$













  • $begingroup$
    That's a good point, imputation could be more robust on changes in the way data is missing. I will take your statement on throwing away past data as an exaggeration though - including a time variable and retraining the model should be enough make it usable again.
    $endgroup$
    – gsmafra
    32 mins ago










  • $begingroup$
    @gsmafra: In general, I don't think adding a time variable will fix the problem. For example, in a random forest, the time variable will only be included in 1/3 of the trees, so it won't even be included in the majority of the decision trees in your random forest.
    $endgroup$
    – Cliff AB
    24 mins ago










  • $begingroup$
    To be clear, I don't think you should throw out your data...but I'd only advise doing "impossible value imputation" on variables you don't think will be very predictive to start with or you're fairly certain that the missingness distribution is fairly stable.
    $endgroup$
    – Cliff AB
    23 mins ago
















2












2








2





$begingroup$

One reason you may not want to use "insert impossible value" methods is that means that your predictive model works conditional on the distribution of the data missingness remaining unchanged. Thus, if after building your tree model, it is realized that we can start using certain features more often, we can no longer use the model that was built using the "impute impossible value" method without retraining the model.



In fact, this problem is even further compounded if the rates of missingness changes during the data collection process itself. Then, even immediately after building the model, it is already "out of date", as the current rates of missingness will be different than the rates of missingness during when the data was collected.



To illustrate the issue, let's suppose a bank is building a database to help predict if clients will default on a loan. Early in the data collection process, loan officers have the option to conduct a background investigation, but they almost never do for clients they deem as trustworthy. Thus, for the especially trustworthy customers, the background check variable is almost always missing. If you use the "impute impossible value" method, having a possible value for background checks indicates high risk.



If background check rates don't change at all, then this "impute impossible value" method will likely still provide valid predictions. However, let's suppose the bank realizes that background checks are really helpful for assessing risk, so they change their policy to include background checks for everyone. Then, everyone will have a possible value for background checks and using the "impute impossible value" method, everyone will be flagged as "high risk".



Cross validation will not catch this issue, as the missingness distribution will be the same between the training and testing sets. So even though the "impute impossible value" method may lead to pretty results during cross-validation, this will lead to poor predictions upon deployment!



Note that you will essentially need to throw away all your data everytime your data collection policy changes! Alternatively, if you can correctly impute the missing values and their uncertainty, you can now use the data that was collected under the old policy.






share|cite|improve this answer











$endgroup$



One reason you may not want to use "insert impossible value" methods is that means that your predictive model works conditional on the distribution of the data missingness remaining unchanged. Thus, if after building your tree model, it is realized that we can start using certain features more often, we can no longer use the model that was built using the "impute impossible value" method without retraining the model.



In fact, this problem is even further compounded if the rates of missingness changes during the data collection process itself. Then, even immediately after building the model, it is already "out of date", as the current rates of missingness will be different than the rates of missingness during when the data was collected.



To illustrate the issue, let's suppose a bank is building a database to help predict if clients will default on a loan. Early in the data collection process, loan officers have the option to conduct a background investigation, but they almost never do for clients they deem as trustworthy. Thus, for the especially trustworthy customers, the background check variable is almost always missing. If you use the "impute impossible value" method, having a possible value for background checks indicates high risk.



If background check rates don't change at all, then this "impute impossible value" method will likely still provide valid predictions. However, let's suppose the bank realizes that background checks are really helpful for assessing risk, so they change their policy to include background checks for everyone. Then, everyone will have a possible value for background checks and using the "impute impossible value" method, everyone will be flagged as "high risk".



Cross validation will not catch this issue, as the missingness distribution will be the same between the training and testing sets. So even though the "impute impossible value" method may lead to pretty results during cross-validation, this will lead to poor predictions upon deployment!



Note that you will essentially need to throw away all your data everytime your data collection policy changes! Alternatively, if you can correctly impute the missing values and their uncertainty, you can now use the data that was collected under the old policy.







share|cite|improve this answer














share|cite|improve this answer



share|cite|improve this answer








edited 47 mins ago

























answered 1 hour ago









Cliff ABCliff AB

13.5k12567




13.5k12567












  • $begingroup$
    That's a good point, imputation could be more robust on changes in the way data is missing. I will take your statement on throwing away past data as an exaggeration though - including a time variable and retraining the model should be enough make it usable again.
    $endgroup$
    – gsmafra
    32 mins ago










  • $begingroup$
    @gsmafra: In general, I don't think adding a time variable will fix the problem. For example, in a random forest, the time variable will only be included in 1/3 of the trees, so it won't even be included in the majority of the decision trees in your random forest.
    $endgroup$
    – Cliff AB
    24 mins ago










  • $begingroup$
    To be clear, I don't think you should throw out your data...but I'd only advise doing "impossible value imputation" on variables you don't think will be very predictive to start with or you're fairly certain that the missingness distribution is fairly stable.
    $endgroup$
    – Cliff AB
    23 mins ago




















  • $begingroup$
    That's a good point, imputation could be more robust on changes in the way data is missing. I will take your statement on throwing away past data as an exaggeration though - including a time variable and retraining the model should be enough make it usable again.
    $endgroup$
    – gsmafra
    32 mins ago










  • $begingroup$
    @gsmafra: In general, I don't think adding a time variable will fix the problem. For example, in a random forest, the time variable will only be included in 1/3 of the trees, so it won't even be included in the majority of the decision trees in your random forest.
    $endgroup$
    – Cliff AB
    24 mins ago










  • $begingroup$
    To be clear, I don't think you should throw out your data...but I'd only advise doing "impossible value imputation" on variables you don't think will be very predictive to start with or you're fairly certain that the missingness distribution is fairly stable.
    $endgroup$
    – Cliff AB
    23 mins ago


















$begingroup$
That's a good point, imputation could be more robust on changes in the way data is missing. I will take your statement on throwing away past data as an exaggeration though - including a time variable and retraining the model should be enough make it usable again.
$endgroup$
– gsmafra
32 mins ago




$begingroup$
That's a good point, imputation could be more robust on changes in the way data is missing. I will take your statement on throwing away past data as an exaggeration though - including a time variable and retraining the model should be enough make it usable again.
$endgroup$
– gsmafra
32 mins ago












$begingroup$
@gsmafra: In general, I don't think adding a time variable will fix the problem. For example, in a random forest, the time variable will only be included in 1/3 of the trees, so it won't even be included in the majority of the decision trees in your random forest.
$endgroup$
– Cliff AB
24 mins ago




$begingroup$
@gsmafra: In general, I don't think adding a time variable will fix the problem. For example, in a random forest, the time variable will only be included in 1/3 of the trees, so it won't even be included in the majority of the decision trees in your random forest.
$endgroup$
– Cliff AB
24 mins ago












$begingroup$
To be clear, I don't think you should throw out your data...but I'd only advise doing "impossible value imputation" on variables you don't think will be very predictive to start with or you're fairly certain that the missingness distribution is fairly stable.
$endgroup$
– Cliff AB
23 mins ago






$begingroup$
To be clear, I don't think you should throw out your data...but I'd only advise doing "impossible value imputation" on variables you don't think will be very predictive to start with or you're fairly certain that the missingness distribution is fairly stable.
$endgroup$
– Cliff AB
23 mins ago




















draft saved

draft discarded




















































Thanks for contributing an answer to Cross Validated!


  • Please be sure to answer the question. Provide details and share your research!

But avoid



  • Asking for help, clarification, or responding to other answers.

  • Making statements based on opinion; back them up with references or personal experience.


Use MathJax to format equations. MathJax reference.


To learn more, see our tips on writing great answers.




draft saved


draft discarded














StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstats.stackexchange.com%2fquestions%2f397942%2fis-it-ever-recommended-to-use-mean-multiple-imputation-when-using-tree-based-pre%23new-answer', 'question_page');
}
);

Post as a guest















Required, but never shown





















































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown

































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown







Popular posts from this blog

Why do type traits not work with types in namespace scope?What are POD types in C++?Why can templates only be...

Will tsunami waves travel forever if there was no land?Why do tsunami waves begin with the water flowing away...

Simple Scan not detecting my scanner (Brother DCP-7055W)Brother MFC-L2700DW printer can print, can't...