Round down and Round up?

Hi everyone,

I have two numbers and want to round up and round down them.

minval = 0.4410 and maxval=0.8450

I want to round down the first number and it will be 0.4. The 2nd number after being rounded up will be 0.9.

I tried with floor(minval) it returns 0 and ceil(maxval) it returns 1. Those numbers are not the result I want.

Could someone give me some advice?

I'm using MATLAB R2014a.

Thank you so much.

[shaadinama]

In MATLAB, the functions "floor" and "ceil" can be used to round down and round up, respectively.

The "floor" function rounds a given input value down to the nearest integer. For example, "floor(3.7)" would return 3, and "floor(-2.3)" would return -3.

The "ceil" function, on the other hand, rounds a given input value up to the nearest integer. For example, "ceil(3.2)" would return 4, and "ceil(-2.9)" would return -2.

Both functions can be useful in various mathematical and programming contexts where it is necessary to round numbers to the nearest whole integer.

Answer 1

[Star Strider]

In R2014b, the round function now supports rounding to a specific number of digits:

minval = 0.4410; 
maxval = 0.8450;

r1_minval = round(minval, 1);
r1_maxval = round(maxval, 1);

If you don’t have R2014b, this works:

diground = @(x,d) round(x*10^d)/10^d;

r2_minval = diground(minval,1);
r2_maxval = diground(maxval,1);

[Khanh]

Thank you.

[Star Strider]

My pleasure!

[Jan]

@NaySan: Please use flags only, to mark inapproriate messages, which should be reviewed by the admins or editors. Thanks.

[Pablo López]

Hi!

This code rounds down both values instead of rounding maxval to 0.9. Is it possible to specify what we would like to do? (round up or down)

Thanks!

[Star Strider]

Probably. You would have to write your own function.

[marchammer]

If the requirement is based on the 2nd decimal as it could be inferred in the problem statement, given a common input and no prompt as to whether to round up or down, round(round(val,2),1) will work. Otherwise the first answer with floor() and ceil() multiplying by a power of 10 and dividing again after rounding is an easy and common way to do this.

It is interesting that the accepted answer doesn't round the second value up as requested.

[Stephen23]

"It is interesting that the accepted answer doesn't round the second value up as requested. "

It is simply because the actual value is less than 5 at that digit (not equal to 5 as it seems when displayed to a much lower precision):

minval = 0.4410

minval = 0.4410

maxval = 0.8450

maxval = 0.8450

fprintf('%.40f\n',minval,maxval)

0.4410000000000000031086244689504383131862
0.8449999999999999733546474089962430298328

So ROUND is working correctly, it is another reminder to always be aware of the behaviors of floating point numbers.

Given hints in the the OP's question, it seems like the correct answer was given by Rajesh here:

https://www.mathworks.com/matlabcentral/answers/168478-round-down-and-round-up#answer_1224129

Answer 2

[Rajesh]

1. floor(minval*10)/10
2. ceil(maxval*10)/10

[Stephen23]

+1 This actually returns what the OP requested:

minval = 0.4410; 
maxval = 0.8450;
floor(minval*10)/10

ans = 0.4000

ceil(maxval*10)/10

ans = 0.9000