Looking for a good and complete probability and statistics book I never had the  opportunity to visit a stats course from a math faculty. I am looking for a probability theory and statistics book that is complete and self-sufficient. By complete I mean that it contains all the proofs and not just states results. By self-sufficient I mean that I am not required to read another book to be able to understand the book. Of course it can require college level (math student) calculus and linear algebra.
I have looked at multiple books and I didn't like any of them.


*

*DeGroot & Schervish (2011) Probability and Statistics (4th Edition) Pearson
This is not complete enough. It just states a lot of stuff without the derivation.  Besides that I like it.

*Wasserman (2004) All of Statistics: A Concise Course in Statistical Inference Springer.
Didn't like it at all. Almost no explanations.
"Weighing the Odds" from David Williams is more formal than DeGroot and seems to be complete and self-sufficient. However, I find the style strange. He also invents new terms that only he seems to use. All the stuff that is explained in DeGroot too is explained better there.
If you know a great book in German that's also fine as I am German.
 A: If you want to read probability as a story, read the best book ever by Feller. I am also guessing that you do not want to go to the level of measure theoretic definition of probabilities which has specialized books. another beginner level book is from Ross. Other specialized applications have specialized books. so more information will gather better suggestions.
A: I would recommend two books not mentioned, as well as several already mentioned.
The first is E.T. Jaynes "Probability: The Language of Science."  It is polemic and he is a very partisan author, but it is very good.
The second is Leonard Jimmie Savage's "The Foundations of Statistics."  You will probably be very surprised when you first start reading it as you will not expect it to go the route it goes.
Both are writing foundational work in Bayesian probability and Bayesian statistics.  The above works are non-Bayesian.
Both books are completely contained and self-sufficient.  Indeed, they build from the foundation upward.  Both approach it axiomatically.  
A: Finding a single, comprehensive book will be very difficult.  If you're asking because you want to do some self-study, get a couple of used texts instead of a single new one.  You can get classics for $3-10 dollars if you look around online.
Feller's "Introduction to Probability" is great for its completeness and expository style, but I don't like the exercises much.  And the exposition would not make it so good for a reference.  He tends to have a lot of long examples, which is great for fostering understanding, and not so great for looking things up.
I enjoyed Allan Gut's "An Intermediate Course in Probability".  There is some overlap with Feller, but it goes into greater depth on those topics.  He covers the various transformations, order statistics (which, if I recall, Feller only does by example).
Ross' Introduction to Probability Models is pretty comprehensive, but it is very example oriented.  Again, that is not my favorite style (I'd rather they saved those examples for exercises with hints, and kept them out of the main flow), but if it works for you, I can recommend it.
You might as well consider Cacoullos' "Exercises in Probability" and Mosteller's "50 Challenging Exercises in Probability".
A: If you are searching for proofs, I have been working for some time on a free stats textbook that collects lots of proofs of elementary and less elementary facts that are difficult to find in probability and statistics books (because they are scattered here and there). You can have a look at it at http://www.statlect.com/
A: For the probability side I like Probability and Random Processes by Grimmett & Stirzaker.  It has a nice way of giving intuitive explanations whilst still being fairly rigorous and providing some proofs at least.
For the Statistics side I've had Theory of Statistics by Schervish on my wish list for a while now but not got around to buying it, so I can only say I've heard good things about it...it's supposed to be a graduate level introduction so possibly more rigorous than the other Schervish book you mention.
A: I recommend Probability Theory and Mathematical Statistics by Marek Fisz, because:


*

*It contains most of the common proof, but without making the book too difficult as an introduction book

*It is quite theoretical, but still contain enough well-designed examples to illustrate points 

*Exercises are meaningful. Some of them are more advanced famous
results

A: As noted by many others, there is no single good text for any scientific subject simply because any given authors or group of authors use a set of assumptions regarding the readers' level of understanding and diversity of knowns and unknowns in the user's brain. Said this, my suggestion for someone knows basics in calculus and linear algebra is to begin with the "modern mathematical statistics with applications" by Devore and Berk. 
A: You can read Student's Solutions Guide for Introduction to Probability, Statistics, and Random Processes book. It provides clear examples and exercises with "additional questions" at the end of each chapter which really help improve learning and there is a logical progression from one idea to another.
