Bankers Rounding
-
Actually, this :
You are a brainless moron.
Please don't take it as an insult, it's just an observation on my part.
is an insult no matter how you put it. I am also easily a type a personality when it comes to my work. I can be very opinionated. I, however, will admit when I am wrong or have made a mistake.
You have not proved to me that there is not a case for using a Bankers Round function. I have used it in a business environment, with the approval and acceptance of the accounting department and the auditors. This means there is a valid case for using it.
-
>> If you wish to continue the academic argument, please find someone else who is interested in such an argument.
You mean who is smart enough and can understand this stuff?
Actually, I mean someone who really cares about having an academic discussion about this. I am quite intelligent enough to to understand this, in fact I have a minor in mathematics, and I fully comprehend your arguements and would actually support it in an academic discussion. We, are however, talking about the real world of business, and there are times when for example, the following is true:
round(TotalSalesAmt * TaxRate, 2) < (round(TotalSalesAmt * TaxRate1, 2) + round(TotalSalesAmt * TaxRate2, 2) + round(TotalSalesAmt * TaxRate2, 2)) where TaxRate = TaxRate1 + TaxRate2 + TaxRate3
This is the case where a bankers round function helps, because the only time I have had this fail was when the some of the components was LESS THAN the total tax rate, and then it was off by 0.01, and we simply applied a business rule to assign the extra penny.
-
This is the case where a bankers round function helps, because the only time I have had this fail was when the some of the components was LESS THAN the total tax rate, and then it was off by 0.01, and we simply applied a business rule to assign the extra penny.
So, you admit: there are some cases when Bankers Rounding does not work. And you need to apply some work around to fix its results.
That means it cannot be used in this type of calculations.
Any example where Banking Rounding works?
_____________
Code for TallyGenerator -
>> You have not proved to me that there is not a case for using a Bankers Round function.
This clearly illustrates your "minor in math".
Non-existence should not be proved.
If there are no cases when this should be used then it should never been used.
It's some essential logic. Applied everywhere.
You should not prove you did not commit a crime.
They must prove you did.
If they cannot provide an evidence - there are no evidences.
If you cannot provide a case where Banking Rounding is a solution - there are no cases.
_____________
Code for TallyGenerator -
And do you admit that the standard rounding function also fails on occasion (look at my example)? You would have to apply "some work around to fix its results" (your own words, by the way). Business in not about absolutes, it is about compromises. When dealing with taxing authorities it is ALWAYS best to have to add the penny to a predetermined choice of accounts (a business rule) rather than taking away a penny (although it may still be only a penny).
-
It does not fail if you do it right, as math school book tells:
Do calculations using 4 digits precision (money datatype) and compare displayed values (2 digits precision).
BTW, taxing authorities don't mind you rounding errors if they stay < $1 in total for reporting period.
_____________
Code for TallyGenerator -
This:
You should not prove you did not commit a crime.
They must prove you did.
If they cannot provide an evidence - there are no evidences.
This depends on where you are being tried. Under US law (possibly Australian and New Zealand as well) the prosecution must prove beyond a reasonable doubt (innocent people HAVE been convicted before). Uder the law in Mexico during the time of Texas independance it was up to the DEFENSE that the defendent was innocent.
Again, this has no bearing on the use and validity of the bankers rounding function.
And, by the way, I am very tired of your personal attacks. If you can't support your arguements without them, you must not have much of an argument to begin with. Try being civil in the future, you may find you get a much more interesting arguemtn focused more on the subject, rather than the individuals which is where it should be.
-
I did not start personal attacks.
Don't want to be attacked - do not attack.
_____________
Code for TallyGenerator -
You know Skippy, earlier you were making fun of the folks behind this rounding method missing 5th and 6th grade math. I think the problem is that you remained there, while the rest of the world moved on to adopt whichever algorithm best meets their needs. I honestly don't believe you're this dense, so the only conclusion I can come to is that you have some sort of allergic reaction to being mistaken. I'm sure there are pills for that.
"Convergent Rounding" or "Round to Nearest Even" or "Round Half Even", also called "Banker's rounding", also called "Statistician's rounding" (yep, that's right, statistician's), and many other names is a perfectly valid rounding method, and is considered an important method by IBM, IEEE, Microsoft, about every signal processor manufacturer in the world, the PLC manufacturing industry, and just about anyone else who matters. So, my choices are. Do I believe some anonymous, obstinent person on the Internet who actually provided us a test that ended up showing he was incorrect and that this rounding method performed what it was designed to do flawlessly, or those other guys? I'll give you until midnight (no matter which day you think it falls on) to guess which.
As I noted earlier, debating the merits of the various rounding methods in various applications is a perfectly worthwhile endeavor, as no method is perfect for all situations. In fact, one of the major problems with this method is that it is confusing to those people who haven't had a lot of math exposure. Nevertheless, the method in question is perfectly useful, and is the ideal choice for many people and many situations, especially when bias avoidance is desirous.
So, are you ready to admit you were wrong, or are you going to pretend that you're still correct and ask us another silly, but inapplicable question, such as how old someone is that was born on Leap Day in the year 1960?
-
I made ONE statement about your personality, I did not continously attack you in post after post, and I even apologized in a private email about that statement (something you have NOT done).
The following are all statements made by you through out this thread:
Author of this conception missed something important on 5th or 6th grade in school.
Someone who wrote this definitely missed couple of important things in school.
I stated that whole idea of Bankers Rounding is stupid. And I can prove it.
"Too much proud" problem?
That's where brilliant software built on something like Bankers Rounding fails and the boring academic guy comes to fix it and save the company from Revenue Department charges (before him several people had to go every day through every bloody Invoice Report and manually correct odd numbers).
This clearly illustrates your "minor in math".
If these are not personal attacks against myself or others, I don't know how to classify them. I stand by this statement: you won't admit when you are wrong.
You don't always have 4 decimals to the right of the decimal position because you don't always have access to a relational database with a money datatype. Sometimes you have to work with a COBOL system using amount fields declared as 9(9)V99. You are distributing a total tax amount round(SalesAmt * TaxRate, 2) over multiple taxing authorities such that the sum of the individual calculations is greater the the amount charged (round(SalesAmt * TaxRate, 2)), obviously, the rounding isn't working properly.
The bankers Round function prevents this type of error. It doesn't prevent all rounding errors. The accountants expressed very strongly that it is best to err on the low side and add the extra penny to one of the accounts than err high and have to take it away.
Again, business in the real world is not about absolutes, it is about making compromises. Are you willing to make a compromise?
-
"Convergent Rounding" or "Round to Nearest Even" or "Round Half Even", also called "Banker's rounding", also called "Statistician's rounding" (yep, that's right, statistician's), and many other names is a perfectly valid rounding method, and is considered an important method by IBM, IEEE, Microsoft, about every signal processor manufacturer in the world, the PLC manufacturing industry, and just about anyone else who matters. So, my choices are.
Does the name "Giordano Bruno" say to you anything?
I believe the names "Catholic Church" and "Holy Inquisition" were not less at that times than IBM, IEEE, Microsoft and all bankers all together at present time.
At least they had the authority to burn him for statements contradicted their believes.
It took 400 years for Catholic Church to admit they were wrong and Giordano was right. But it does not change the fact that all big names were wrong.
You are entitled to follow blindly big names, but it does not mean you go in right direction.
So, your reference to big names cannot prove anything.
And you still did not prove that Bankers Rounding is a valid approach.
P.S. I have not found any personal attack on you in the quotes you posted.
Only "minor in math" could be considered as that, but those are your words, not mine.
And if you mean to apologize for something you must do it same way in front of the same audience as you did the action you're apologizing for.
_____________
Code for TallyGenerator -
> The bankers Round function prevents this type of error.
It does not.
It prevents some cases of this type of errors.
Very limited cases.
_____________
Code for TallyGenerator -
Sergiy, it was the way you put it, not the way I put it. As I tell my kids, it is the tone or inclination you put on words as much as the words themselves.
My B.S. degree is a double major in Computer Science and Business Administration with a double minor in Mathemetics and Technical Writing. I also have an M.A. In Computer Information Resource Management and Space Systems Management. I have had over 21 semester hours in Mathematics, so I have a strong understanding in this area.
And, as David McFarland stated, every rounding method has it issues in one way or another. As for continuing this debate, it is futile as the only way it will end is if everyone except you admits they are wrong, and that isn't going to happen.
You won't even accept the fact that your own test proved what was being said. That is the sad part of all of this.
-
Your post:
Does the name "Giordano Bruno" say to you anything?
I believe the names "Catholic Church" and "Holy Inquisition" were not less at that times than IBM, IEEE, Microsoft and all bankers all together at present time.
At least they had the authority to burn him for statements contradicted their believes.
It took 400 years for Catholic Church to admit they were wrong and Giordano was right. But it does not change the fact that all big names were wrong.
Took the time to read a bit about Giordano Bruno, and once again you manage bring up something that has nothing to do with the subject at hand.
-
Once again, for really "smart" people with big diplomas:
I ready to admit anything you can prove.
As soon as you cannot prove validity of Bankers Rounding there is nothing to admit for me.
My statement remains the same: it's absolutely wrong method based on wrong assumptions.
_____________
Code for TallyGenerator
Viewing 15 posts - 76 through 90 (of 373 total)
You must be logged in to reply to this topic. Login to reply