Cox PH model in `lifelines` - violated assumptions for dummy variables - python

I am using lifelines library to estimate Cox PH model. For the regression I have many categorical features, which I one-hot-encode and remove one column per feature to avoid multicollinearity issue (dummy variable trap). I am not attaching the code as the example can be similar to the one given in the documentation here.
By running cph.check_assumptions(data) I receive information that each dummy variable violates the assumptions:
Variable 'dummy_a' failed the non-proportional test: p-value is 0.0063.
Advice: with so few unique values (only 2), you can try `strata=['dummy_a']` in the call in `.fit`. See documentation in link [A] and [B] below.
How should I understand the advice in terms of multiple dummy variables for a single categorical feature? Should I add them all to strata?
I will appreciate any comments :)

#abu, your question brings up a clear gap in the documentation - what to do if dummy variables violate the proportional test. In this case, I suggest not dummying the variable, and add the original column as a stratified variable, ex: fit(..., strata=['dummy'])

Related

How and when to deal with outliers in your dataset (general strategy)

I stumbled about the following problem:
I'm working on a beginners project in data science. I got my test and train data splits and right now I'm analyzing every feature, then adding it to either a dataframe for discretised continuous variables or a dataframe for continuous variables.
Doing so I encountered a feature with big outliers. If I would to delete them, other features I already added to my sub dataframes would have more column entries than this one.
Should I just find a strategy to overwrite the outliers with "better" values or should I reconsider my strategy to split the train data for both types of variables in the beginning? I don't think that
getting rid of the outlier rows in the real train_data would be useful though...
There are many ways to deal with outliers.
In my cours for datascience we used "data imputation":
But before you start to replace or remove data, its important to analyse what difference the outlier makes and if the outlier is valid ofcours.
If the outlier is invalid, you can delete the outlier and use data imputation as explained below.
If your outlier is valid, check the differnce in outcome with and without the outlier. If the difference is very small then there ain't a problem. If the differnce is significant you can use standardization and normalization.
You can replace the outlier with:
a random value (not recommended)
a value based on hueristic logic
a value based on its neighbours
the median, mean or modus.
a value based on interpolation (making a prediction with a certain ml model)
I recommend using the strategy with the best outcome.
Statquest explains datascience and machinelearning concepts in a very easy and understandable way, so refer to him if you encounter more theoritical questions: https://www.youtube.com/user/joshstarmer

How should I handle NaN values in a Finance DF?

I am a beginner in Machine Learning, my point is..how should i encode the column "OECDSTInterbkRate"? I don't know how to replace the missing values and especially with what. Should I just delete them? Or replace them with the mean / median of the values?
There are many approaches to this issue.
The simplest: if you have huge amount of data - drop NaNs.
Replace the NaNs with mean/median/etc of the whole non-NaN dataset or the dataset grouped by one or several columns. E.g. for you dataset you can fill the Australia NaNs with a mean for Australian non-NaNs. And the same for other countries.
A common approach is to create another indicator column after the imputation of NaNs which keeps the indices where the missing data was replaced with a value. This column then is taken as yet another input to your ML algorithm.
Take a look at the docs (assuming you work with Pandas) - the developers of the library have already created some tools for the missing data: https://pandas.pydata.org/pandas-docs/stable/user_guide/missing_data.html
There's no specific answer to your question, it's a general problem in statistics which is called "imputation". Depending on the application the answer could be many things.
There are few alternatives that comes to mind first to solve your problem, but don't forget that "no data" is almost always better than "bad/wrong data". If you have more than enough rows without the rows with NaNs, you may simply drop them. Otherwise you can consider the following:
Can you mathematically calculate the column that you need by the other columns that you already have in your dataset? If so, you have your answer.
Check the correlation of the particular column by using it's non-missing valued rows with the other columns and see if they are highly correlated. If so, you might just as well try dropping the whole column(might not be always a good idea but it's generally a good idea).
Can you create an estimator(such as a regression model) to predict the missing values by learning the pattern using the values that you already have and by using the other columns with a really good accuracy? Well you might have an answer (need benchmarking with the following). Please keep in mind that this is a very risky operation that could give bad estimations and decrease the performance of your overall model. Try this only if your estimations are really good!
Is it a regression problem? Using the statistical mean could be a good idea.
Is it a classification problem? Using median could be a good idea.
In some cases using mode might also be a good idea depending on the distribution.
I suggest that you try all the things out and see which one works better because there's really not a concrete answer to your problem. You can create a machine learning model without using the column and use it's performance as a baseline, and carry out a performance(accuracy) benchmarking for all the steps compared to the baseline.
Note: I am just a graduate student with some insights, please comment out if anything I said is not correct!

Find probaility of failure using logistic regression in python

I am trying to get some insights from a dataset using Logistic Regression. The starting dataframe has the information of if something was removed unscheduled (1 = Yes and 0 = No) and some data which was provided with this removal.
This looks like this:
This data is then 'dummyfied' using pandas.get_dummies, where the result is as expected.
Then I normalized the found coefficients (using coef_) to scale everything the same. I put this in a dataframe with the column 'Parameter' (which is the column name of the dummy dataframe) and the column 'Value' (which is the value obtained using the coefficients).
Now I will get the following result.
Now this result shows that the On-wing Time is the biggest contributor in the unscheduled removal.
Now the question: how can I predict what the chance is that there will be an unscheduled removal for this reason (this column)? So, what is the chance that I will get another unscheduled removal which is caused by the On-wing Time?
Note that these parameters can change since this data is fake data and data may be added later on. So when the biggest contributor changes, the prediction should also focus on the new biggest contributor.
I hope you understand the question.
EDIT
The complete code and dataset (fake one) can be found here: 1drv.ms/u/s!AjkQWQ6EO_fMiSEfu3vYgSTBR0PZ
Ganesh

Missing data in Dataframe using Python

[]
Hi ,
Attached is the data, can you please help me to handle the missing data in the "Outlet_Size" column.
So that i can use this complete data for preparing the datascience models.
Thanks,
These are one of the major challenges of Data Mining problems (or Machine Learning). YOU decide what to do with the missing data based on PURE EXPERIENCE. You mustn't look at Data Science as a blackbox that follows a series of steps to be successful at it!
Some guidelines about missing data.
A. If more than 40% of the data is missing from a column, drop it! (Again, the 40% depends on what type of problem you're working with! If the data is super crucial or its very trivial that you can ignore it).
B. Check if there is someway you can impute the missing data from the internet. You're looking at item weight! If there is anyway you could know which product you're dealing with instead of hashed coded Item_Identifier, then you can always literally Google it and figure it out.
C. Missing data can be classified into two types:
MCAR: missing completely at random. This is the desirable scenario in case of missing data.
MNAR: missing not at random. Missing not at random data is a more serious issue and in this case it might be wise to check the data gathering process further and try to understand why the information is missing. For instance, if most of the people in a survey did not answer a certain question, why did they do that? Was the question unclear?
Assuming data is MCAR, too much missing data can be a problem too. Usually a safe maximum threshold is 5% of the total for large datasets. If missing data for a certain feature or sample is more than 5% then you probably should leave that feature or sample out. We therefore check for features (columns) and samples (rows) where more than 5% of the data is missing using a simple function
D. As posted in the comments, you can simply drop the rows using df.dropna() or fill them with infinity, or fill them with mean using df["value"] = df.groupby("name").transform(lambda x: x.fillna(x.mean()))
This groups the column value from dataframe df by category name, finds the mean in each category and fills the missing value in value with the corresponding mean of that category!
E. Apart from just either dropping missing values, replacing with mean or median, there are other advanced regression techniques you can use that has a way to predict missing values and fill it, E.G (mice: Multivariate Imputation by Chained Equations), you should browse and read more about where advanced imputation technique will be helpful.
The accepted answer is really nice.
In your specific case I'd say either drop the column or assign a new value called Missing. Since that's a Categorical variable, there's a good chance it ends up going into a OneHot or Target Encoder (or being understandable by the model as a category directly). Also, the fact the value is NaN is an info itself, it can come from multiple factors (from bad data to technical difficulties getting an answer, etc). Be careful and watch this doesn't brings bias or some information you shouldn't know (example : the products have NaN due to not being into a certain base, thing that will never happen in a real situation, which will make your result non-representative of a true situation)
The column "Outlet_Size" contains the categorical data, so instead of dropping the data use measures to fill data.
Since it is categorical data use Measures of Central Tendency, Mode.
Use mode to find which category occurs more or frequently and fill the column with the corresponding value.
Code:
Dataframe['Outlet_Size'].mode()
Datarame['Outlet_Size'].fillna(Dataframe['Outlet_Size'].mode(), inplace=True)

Handling unassigned (null) values of features in regression (machine learning)?

I want to do linear regression analysis. I have multiple features. Some features has unassigned (null) values for some items in data. Because for some items some specific feature values were missed in data source. To be more clear, I provide example:
As you can see, some items missing values for some features. For now, I just assigned it to 'Null', but how to handle this values when doing linear regression analysis of the data? I do not want this unassigned values to incorrectly affect regression model. Unfortunately I cannot get rid of items where unassigned feature values presented. I plan to use Python for regression.
You need to either ignore those rows -- you've already said you can't, and it's not a good idea with the quantity of missing values -- or use an algorithm that proactively discounts those items, or impute (that's the technical term for filling in an educated guess) the missing data.
There's a limited amount of help we can give, because you haven't given us the semantics you want for missing data. You can impute some of the missing values by using your favourite "closest match" algorithm against the data you do have. For instance, you may well be able to infer a good guess for area from the other data.
For your non-linear, discrete items (i.e. District), you may well want to to keep NULL as a separate district. If you have few enough missing entries, you'll be able to get a decent model anyway.
A simple imputation is to replace each NULL with the mean value for the feature, but this works only for those with a proper mean (i.e. not District).
Overall, I suggest that you search for appropriate references on "impute missing data". Since we're not sure of your needs, we can't help much with this, and doing so is outside the scope of SO.

Categories

Resources