write a R code for Portfolio Optimisation using genetic algorithm using R language using Rmarkdown format: Part $1$

The focus of the first part of the assignment is on the optimisation of the portfolio using a GA$.$ To begin with, manually choose a set of assets to work with $-$ which ones you select are entirely up to you, but aim to work with around $10$ and also try to pick ones which don't all exhibit the same behaviour $($e$.$g$.$ maybe take them from different sectors$).$

Then use a genetic algorithm $($and the GA package$)$ to identify the optimal set of weights for your portfolio which gives the best balance of risk versus return.

Having done this you then need to conduct a detailed evaluation of the portfolio by looking at how well it might perform in the "future", and comparing it with other possible portfolios $($particularly a balanced portfolio, where all weights are even, and a set of randomly generated portfolios$).$

Finally $($for this part$)$ try exploring different weightings of risk and return. The idea behind this it to try and emulate a multi$-$objective approach and you will need to think about how it is possible to modify the fitness function to create differently balanced portfolios $-$ e$.$g$.$ at the extremes might be ones which only care about maximising returns or are very risk averse, and in between might be ones which prioritise risk over return and vice$-$versa. How do the weights and the performance of these portfolios compare with each other and your other solutions? I want to use these stocks symbols$<-$ c$("$KFH$.$KW$","$NBK$.$KW$",$ "AGLTY.KW$",$ "ZAIN.KW$",$ "BOUBYAN.KW$",$

"WARBABANK.KW$",$ "OOREDOO.KW$","$STC$.$KW$",$ "NAPESCO.KW$",$ "ABAR.KW$")$ getSymbols$($symbols$,$ src $=$ "yahoo" $,$ from $="2022-01-01",$ to $="2023-01-1")$

ReturnData $<-$ data.frame$($merge$($dailyReturn$($KFH$.$KW$),$

dailyReturn$($NBK$.$KW$),$ dailyReturn$($AGLTY$.$KW$),$ dailyReturn$($ZAIN$.$KW$),$

dailyReturn$($BOUBYAN$.$KW$),$dailyReturn$($WARBABANK$.$KW$),$

dailyReturn$($OOREDOO$.$KW$),$ dailyReturn$($STC$.$KW$),$ dailyReturn$($NAPESCO$.$KW$),$

dailyReturn$($ABAR$.$KW$)))$ meanofReturns $<-$ apply$($ReturnData$,2,$mean$)$ covOfReturn $<-$ cov$($ReturnData$)$ I did only this for part one and I don't know what fitness function to use and how to do the rest of part one and two. Part $2$

Once you have done this successfully, then try extending your solution to select the set of assets for your GA to work on$.$ The pool from which you select should be reasonably large $($minimum of $50)$ and to begin with you may identify the selection criteria so that this GA operates independently of the one in part $1($but would pass on the set of identified assets to be weighted by the GA from part $1).$ Once you have this working $($and if you are feeling ambitious!$)$ then investigate how these might work together $($working out an efficient evaluation mechanism is the key to this part$).$