The results of the numerical analysis programs facilitate scientific research.

Statistics of data ( p value etc.) is very important in scientific research

Of course the language of mathematics is superior to the art language specially in the scientific research which gets more benefits from numerical analysis.

Numerical analysis with its statistical power bolsters the validity of the findings in scientific research.

Previously, result in only % or mean with Standard Error worked for publication in good journals. But now a days no good journal will accept any article without other parameters like P value , Superscription of letter for denoting similarity etc. in data table and analysis.

Many software are there for such works (like SPSS).

We have to accept the change, as technology advances everyday!

Numerical analysis programs are the basis for analyzing the data and results of engineering research, on the basis of which the results of the evaluation and conclusions of the research are based.

thanks for all

best wishes

The accuracy depends on the problem conditions, type of equations, the selected method and algorithem, the sophistication of the computer and platform. many variables are involved, and still some times it is the only resort.

Best regards.

The numerical analysis is much important because its facilate to understanding the idea of the research and its consider the base on witch the conclusions of the scientific research build on ,, Best regards ,, Jawad ,Ali

Good question.

dear

so thanks for all

best wishes

good question

thanks doctor Massar Saieb Kadhim

best regards

Dear Dr....

The numerical analysis is much important because It shows the effectiveness of search results ....

thanks for sharing

thanks doctor Dhia Hussain Jassim Al-Delemi

I agree with you,

best regards

yes , Of course

dear doctor Dhia Hussain Jassim Al-Delemi

best wishes

Good question

thanks doctor Hamid Al-jameel

best regards

It depends on the scientific research that we are doing. The effect of numerical analysis programs is for some research essential because it is the only way for doing scientific research. It can be just an illustration of the results obtained analytically. So, the programs can be considered if there is no analytical way to get a result or in order to validate analytical results.

I recommend this article:

Zhitnikov V. P., Sherykhalina N.M., Sokolova A.A. Problem of Reliability Justification of Computation Error Estimates. Mediterranean Journal of Social Sciences, 2015, Vol. 6, No. 2, pp. 65 – 78. Doi:10.5901/mjss.2015.v6n2s4p65.

so thanks dear

Vladimir Zhitnikov

Hassane Bouzahir

best regards

Especially in the study of multi-extremal problems in which the objective function can be represented only algorithmically the only tool for carrying out the study is a program of numerical analysis (I am sorry for bad English).

THANKS DOCTOR Bahtiyar Bayraktar

HAPPY NEW YEAR

Thank you. Happy New Year and success.

For many practical mathematical modelling problems, for example,

weather forecasting, design of aircraft, testing structures

to be built in an earthquake zone, etc. analytical solutions

to the resultant modelling equations do not exist. Indeed,

for some very simple problems, for example, the evaluation

of definite integrals and the solution of a general single

non-linear equation, there is no simple form function that

particular values of the free parameters (e.g., the limits of

the integration) can be plugged into to obtain a resultant

value. In such circumstances we are forced to use numerical

approximations to obtain our `solutions'. In general numerical

analysis/algorithms/software provides us with the ability

to generate a specific solution to a specific problem; i.e.,

we provide numerical values that define a particular problem

(generally just one of an infinite number of possibilities)

and the numerical software returns a numerical approximation

to his problem. For example, you provide the numerical values

for the coefficient matrix and the right hand side vector

that defines a system of linear equations and the software

will return a vector of numbers that are an approximation to

the solution vector. If we change one of the model parameters

(e.g., one of the matrix coefficients) we have to repeat the

computation to generate a further numerical solution.

The skill in producing numerical software is being able to

generate algorithms and implementations of algorithms (code)

that solves a general class of mathematical problems to a

guaranteed accuracy. Typically a user will have a desired

accuracy in mind which is problem dependent and which could

depend on, for example, the accuracy of the problem defining

parameters; the user then expects the results returned by the

software to be within that tolerance. In most situations the

higher the required accuracy the longer the computed solution

takes to compute; hence, asking for low accuracy where this

is acceptable can save resources.

It is also useful if the software is able to detect problems

that it is not designed to solve, for example, in the

solution of linear equations if the software is not designed

to detect coefficient matrices that are `close to singular'

then misleading results may be generated. (For more details

of this problem look at condition number estimation in Lapack.)

There is an `art' to generating numerical algorithms and

implementing them; there is also an art to using them. Users

who have little or no experience in numerical problem solving

need to tread carefully. Certainly such users should not,

in general, be writing their ow codes. For a multitude of

practical problems, off the shelf software is available and

should be used -- with caution! Try out the software on simple

problems to start with to gain experience in setting up the

user defined parameters; only when you feel confident that you

understand how to use the software correctly should you move

to more complex (useful) problems. Only ever think of writing

your own when all else fails.

Many practical problems can be solved quickly and easily using

freely available numerical packages/libraries. However, you,

the user, are responsible for checking that the results are

plausible, correct and (hopefully) a good approximation to

the problem you originally wanted to solve.

So the answer to your original question is that if used correctly and effectively numerical analysis programs are extremely important to scientific research as they are often the only means available to gain any insight into a multitude of important problems.

THANKS DOCTOR

HAPPY NEW YEAR

I think numerical analysis plays a key role on scientific research. Theory guides practice and gives a good direction of scientific research.

Without numerical analysis, scientific research may be blind. But numerical analysis can be difficult.

Dear Doctor…

Thanks for all…

Appreciate your efforts… best wishes