I want to reduce accuracy and precision to 3 digits.

for example I want to a=2.18+0.139; with 3 digits answer must be 2.31. is any way to do this?

[Tom Wenk]

Do you want to use 2.31 in further calculations or do you want to display the number?

[meh hadi]

in further calculatins

[Tom Wenk]

Then I think you have to use the symbolic toolbox like Walter Robinson stated.

But why do you want to change 2.319 to 2.31 rather than 2.32? If you use this value for further calculations you should round in my opinion..

Answer 1

[Praveen Iyyappan Valsala]

I don't know any inbulit function which can do that. But, you can try something like below

trun=@(num,Ndigits)floor(a*10^Ndigits)/10^Ndigits;
trun_a=trun(a,2);

Answer 2

[Walter Roberson]

No, there is no built-in way.

You can use the symbolic toolbox with digits set to 3, but that will round rather than truncate.

[Stephen23]

Note that setting the digits does not mean that calculations will be performed with exactly that precision, because calculations may use guard digits.

Answer 3

[Tom Wenk]

Like Walter Roberson stated, I think you have to use the symbolic toolbox for that.

Maybe this function from the toolbox can help you: https://de.mathworks.com/help/symbolic/digits.html

[Stephen23]

Note that setting the digits does not mean that calculations will be performed with exactly that precision, because calculations may use guard digits.

[Tom Wenk]

I see, but is there an other way to do it then?

[Walter Roberson]

For addition: multiply the original values by 100, take floor() or fix(). Do the addition. Divide the result by 100.