My data has a lot of missing values and I have to predict those values. One way is to take the average of those values. But I want to hear an other perspective on it. How experienced data scientist solve such kind of issue?
Are your missing values categorical or continuous?
One way is to remove the samples entirely, however this may lead to a sampling bias, since the missing values could have been the result of some causal effect, that is the missing values are not missing completely at random.
If your data has enough dimensionality, you can treat your missing values as the output and try to apply a predicting model and hope that it can faithfully estimate the missing values, given the explanatory variables you already have.
Picking the most frequent value, the median, or averaging as you point out could also be an option, however be careful with outliers when averaging as these can have a tremendous effect on the mean.
It depends on nature of variables, it may be some statistics like mean or median. Another practice is assign to missing variables some value different from others for example 0, -1 or something like this.
The hardest approach is to impute the dataset and not deviate too far from the truth. A test to validate how well you have done this is the following. If the other parameters provide enough evidenced insight to impute with a level of precision for missing data....it should be able to do it with existing data.
So if 60 percent of the column is missing, take the row observations where this column is PRESENT.
Next, randomly choose to remove 60% of this subsetted data. Now run imputation methods of your choosing.
Compare the imputed dataset to the real data set for similarity. Decide if they are close enough for you to then run this against the full data set. At least this approach will give you a leg to stand on if you need to defend yourself.
Fight the Good Fight.
Related
Firstly, I am totally out of my expertise zone so please bear with me.
I developed a fluid dynamic engine with 5 exposed parameters (say A,B,C,D,E). When you give this engine these 5 parameters, it does magic and give out a value 'Z'.
I want to write a script which can explore which combinations of A-E give lowest (or close to lowest) value of Z.
I know optimization algorithm exists, but from all of my search for examples, they use some function.
So I guess my function would simply be minimize Z? But where do A-E go?
Not really an answer, but some questions and ideas that might help you think through the best way to address this. We have no understanding of how big a range of values needs to be explored for those parameters, or how Z behaves, so this is very vague...
If you look at the values of Z for given values of A...E, does the value of Z jump around a lot for small changes on the parameter values, or does the Z value change reasonably smoothly?
If the Z value is not too eratic you could try some kind of gradient descent approach using calculated values of Z for some values of the parameters to approximate the gradient - suppose changing the value of 'A' from 1 to 2 gives a better change in the Z value than a similar size change in the other parameters, then try other values of A while keeping the other parameters fixed until you find a value of A that gives the best value of Z. Then try changing the other parameter values to see which one gives the steepest descent and try to find some better value for that parameter. Repeat this process until you can't find any improvement and you will have found a (local) minimum. You could then start at a different place in your parameter space and try again - you will probably find several local minima, and may just choose the best of those. Not provably optimal but may be good enough. Of course you can get clever and use things like conjugate gradients, Newton-Raphson or similar if Z is smooth enough.
If the Z values are very eratic, then you might have to just do some sampling of the possible combinations of A...E to get values of Z and choose the best you can find. Again you might do that in some systematic way (e.g. points on a grid in your parameter space) or entirely at random, or a combination of both.
If you find that there are 'clusters' of good solutions with similar values of the parameters then maybe some kind of local search would help - the idea is that there is often a better solution in the local neighbourhood of a known good solution. So maybe try perturbing your parameter values a bit from a known solution to see if that can lead to a better solution - either by some gradient descent method or by random sampling.
Unfortunately, if your Z calculation is complex, then any method using it as a black box will likely be slow as it will need to be re-evaluated many times.
You could use a Genetic Algorithm, where your chromosomes are formed with the 5 candidate values of the variables you have to optimize, to minimize Z, and your optimization/fitness "function" is the simulation itself outputting Z.
Other viable alternatives are Particle Swarm Optimization algorithm or Ant Colony Optimization. All of those are usable algortihms for that kind of optimization problem.
I have a tricky question someone asked me:
I got a bunch of columns with data to predict some future sales.
There are a bunch of these columns that use a lot of memory and only got zeros. The question is: Can I just remove these columns from my analysis?
Second part. What if the columns that only have zeros also have missing values. What do you do?
Analyzing datasets with large numbers of zeros can indeed be a waste of computational resources (see sparse matrices computation), especially since they may not be contributing any meaningful information. In fact, they might even add noise to your dataset, obscuring any relationships you might find otherwise.
But there are cases were zeros can be incredibly meaningful. For example, if you were trying to predict future sales of products based on units sold (count data), with each column representing a month of sales, you might want to keep the zeros as they provide insight into sales of your product during those specific months.
Removing missing values is definitely tricky, and they can often be a sign that you might need to re-examine the data collection process for explanation; is there a reason why the data is missing, or does the missing value mean something (e.g. sometimes zeros can be coded as NAs)?
How you deal with zeros/missing values varies greatly based on the domain you're in, the specific question you're trying to answer, and what exactly the meaning of your data/columns are. So it's kind of hard to answer without knowing much about the data itself.
and thanks for reading my thread.
I have read some of the previous posts on formatting/normalising input data for a Neural Network, but cannot find something that addresses my queries specifically. I apologise for the long post.
I am attempting to build a radial basis function network for analysing horse racing data. I realise that this has been done before, but the data that I have is "special" and I have a keen interest in racing/sportsbetting/programming so would like to give it a shot!
Whilst I think I understand the principles for the RBFN itself, I am having some trouble understanding the normalisation/formatting/scaling of the input data so that it is presented in a "sensible manner" for the network, and I am not sure how I should formulate the output target values.
For example, in my data I look at the "Class change", which compares the class of race that the horse is running in now compared to the race before, and can have a value between -5 and +5. I expect that I need to rescale these to between -1 and +1 (right?!), but I have noticed that many more runners have a class change of 1, 0 or -1 than any other value, so I am worried about "over-representation". It is not possible to gather more data for the higher/lower class changes because thats just 'the way the data comes'. Would it be best to use the data as-is after scaling, or should I trim extreme values, or something else?
Similarly, there are "continuous" inputs - like the "Days Since Last Run". It can have a value between 1 and about 1000, but values in the range of 10-40 vastly dominate. I was going to scale these values to be between 0 and 1, but even if I trim the most extreme values before scaling, I am still going to have a huge representation of a certain range - is this going to cause me an issue? How are problems like this usually dealt with?
Finally, I am having trouble understanding how to present the "target" values for training to the network. My existing results data has the "win/lose" (0 or 1?) and the odds at which the runner won or lost. If I just use the "win/lose", it treats all wins and loses the same when really they're not - I would be quite happy with a network that ignored all the small winners but was highly profitable from picking 10-1 shots. Similarly, a network could be forgiven for "losing" on a 20-1 shot but losing a bet at 2/5 would be a bad loss. I considered making the results (+1 * odds) for a winner and (-1 / odds) for a loser to capture the issue above, but this will mean that my results are not a continuous function as there will be a "discontinuity" between short price winners and short price losers.
Should I have two outputs to cover this - one for bet/no bet, and another for "stake"?
I am sorry for the flood of questions and the long post, but this would really help me set off on the right track.
Thank you for any help anyone can offer me!
Kind regards,
Paul
The documentation that came with your RBFN is a good starting point to answer some of these questions.
Trimming data aka "clamping" or "winsorizing" is something I use for similar data. For example "days since last run" for a horse could be anything from just one day to several years but tends to centre in the region of 20 to 30 days. Some experts use a figure of say 63 days to indicate a "spell" so you could have an indicator variable like "> 63 =1 else 0" for example. One clue is to look at outliers say the upper or lower 5% of any variable and clamp these.
If you use odds/dividends anywhere make sure you use the probabilities ie 1/(odds+1) and a useful idea is to normalize these to 100%.
The odds or parimutual prices tend to swamp other predictors so one technique is to develop separate models, one for the market variables (the market model) and another for the non-market variables (often called the "fundamental" model).
Many of the variables in the data I use on a daily basis have blank fields, some of which, have meaning (ex. A blank response for a variable dealing with the ratio of satisfactory accounts to toal accounts, thus the individual does not have any accounts if they do not have a response in this column, whereas a response of 0 means the individual has no satisfactory accounts).
Currently, these records do not get included into logistic regression analyses as they have missing values for one or more fields. Is there a way to include these records into a logistic regression model?
I am aware that I can assign these blank fields with a value that is not in the range of the data (ex. if we go back to the above ratio variable, we could use 9999 or -1 as these values are not included in the range of a ratio variable (0 to 1)). I am just curious to know if there is a more appropriate way of going about this. Any help is greatly appreciated! Thanks!
You can impute values for the missing fields, subject to logical restrictions on your experimental design and the fact that it will weaken the power of your experiment some relative to having the same experiment with no missing values.
SAS offers a few ways to do this. The simplest is to use PROC MI and PROC MIANALYZE, but even those are certainly not a simple matter of plugging a few numbers in. See this page for more information. Ultimately this is probably a better question for Cross-Validated at least until you have figured out the experimental design issues.
I have a table representing standards of alloys. The standard is partly based on the chemical composition of the alloys. The composition is presented in percentages. The percentage is determined by a chemical composition test. Sample data.
But sometimes, the lab cannot measure below a certain percentage. So they indicate that the element is present, but the percentage is less than they can measure.
I was confused how to accurately store such a number in an SQL database. I thought to store the number with a negative sign. No element can have a negative composition of course, but i can interpret this as less than the specified value. Or option is to add another column for each element!! The latter option i really don't like.
Any other ideas? It's a small issue if you think about it, but i think a crowd is always wiser. Somebody might have a neater solution.
Question updated:
Thanks for all the replies.
The test results come from different labs, so there is no common lower bound.
The when the percentage of Titanium is less than <0.0004 for example, the number is still important, only the formula will differ slightly in this case.
Hence the value cannot be stored as NULL, and i don't know the lower bound for all values.
Tricky one.
Another possibility i thought of is to store it as a string. Any other ideas?
What you're talking about is a sentinel value. It's a common technique. Strings in most languages after all use 0 as a sentinel end-of-string value. You can do that. You just need to find a number that makes sense and isn't used for anything else. Many string functions will return -1 to indicate what you're looking for isn't there.
0 might work because if the element isn't there there shouldn't even be a record. You also face the problem that it might be mistaken for actually meaning 0. -1 is another option. It doesn't have that same problem obviously.
Another column to indicate if the amount is measurable or not is also a viable option. The case for this one becomes stronger if you need to store different categories of trace elements (eg <1%, <0.1%, <0.01%, etc). Storing the negative of those numbers seems a bit hacky to me.
You could just store it as NULL, meaning that the value exists but is undefined.
Any arithmetic operation with a NULL yields a NULL.
Division by NULL is safe.
NULL's are ignored by the aggregation functions, so queries like these:
SELECT SUM(metal_percent), COUNT(metal_percent)
FROM alloys
GROUP BY
metal
will give you the sum and the count of the actual, defined values, not taking the unfilled values into account.
I would use a threshold value which is at least one significant digit smaller than your smallest expected value. This way you can logically say that any value less than say 0.01, can be presented to you application as a "trace" amount. This remains easy to understand and gives you flexibility in determining where your threshold should lie.
Since the constraints of the values are well defined (cannot have negative composition), I would go for the "negative value to indicate less-than" approach. As long as this use of such sentinel values are sufficiently documented, it should be reasonably easy to implement and maintain.
An alternative but similar method would be to add 100 to the values, assuming that you can't get more than 100%. So <0.001 becomes 100.001.
I would have a table modeling the certificate, in a one to many relation with another table, storing the values for elements. Then, I would still have the elements table containing the value in one column and a flag (less than) as a separate column.
Draft:
create table CERTIFICATES
(
PK_ID integer,
NAME varchar(128)
)
create table ELEMENTS
(
ELEMENT_ID varchar(2),
CERTIFICATE_ID integer,
CONCENTRATION number,
MEASURABLE integer
)
Depending on the database engine you're using, the types of the columns may vary.
Why not add another column to store whether or not its a trace amount
This will allow you to to save the amount that the trace is less than too
Since there is no common lowest threshold value and NULL is not acceptable, the cleanest solution now is to have a marker column which indicates whether there is a quantifiable amount or a trace amount present. A value of "Trace" would indicate to anybody reading the raw data that only a trace amount was present. A value of "Quantity" would indicate that you should check an amount column to find the actual quantity present.
I would have to warn against storing numerical values as strings. It will inevitably add additional pain, since you now lose the assertions a strong type definition gives you. When your application consumes the values in that column, it has to read the string to determine whether it's a sentinel value, a numeric value or simply some other string it can't interpret. Trying to handle data conversion errors at this point in your application is something I'm sure you don't want to be doing.
Another field seems like the way to go; call it 'MinMeasurablePercent'.