R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures...
Transcript of R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures...
![Page 1: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/1.jpg)
R: Programming
140.776 Statistical Computing
September 8, 2011
140.776 Statistical Computing R: Programming
![Page 2: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/2.jpg)
Why programming?
−10 −5 0 5 10
−10
−5
05
10
u
v ● ●
● ●
140.776 Statistical Computing R: Programming
![Page 3: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/3.jpg)
Why programming
−10 −5 0 5 10
−10
−5
05
10
u
v
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
140.776 Statistical Computing R: Programming
![Page 4: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/4.jpg)
Why programming
−10 −5 0 5 10
−10
−5
05
10
u
v
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●● ●
● ●● ●
● ●● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●
● ●● ●
● ● ● ●
● ●
140.776 Statistical Computing R: Programming
![Page 5: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/5.jpg)
Control structures
Programming is more than just putting commands you’ve learnt sofar into a *.R file. A key element of programming (which is alsotrue for other languages) is that you can use control structures tocontrol the flow of execution of the program.
For example, “for()” is a control structure in R to repeatedlyexecute a series of similar commands.
140.776 Statistical Computing R: Programming
![Page 6: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/6.jpg)
Control structures
Control structures commonly used in R include:
if, else: testing a condition
for: execute a loop for a fixed number of times
while: execute a loop while a condition is true
repeat: execute a loop until seeing a break
break: break the execution of a loop
next: skip an iteration of a loop
return: exit a function
140.776 Statistical Computing R: Programming
![Page 7: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/7.jpg)
Conditional execution: if statements
if(<condition>) {## do something
} else {## do something else
}
if(<condition1>) {## do something
} else if (<condition2>) {## do something different
} else {## do something else
}
140.776 Statistical Computing R: Programming
![Page 8: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/8.jpg)
Conditional execution: if statements
Example: compute the absolute value of x and assign it to y.
if(x<0) {y<-(-x)
} else {y<-x
}
140.776 Statistical Computing R: Programming
![Page 9: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/9.jpg)
Conditional execution: if statements
The else clause is not necessary:
if(<condition1>) {## do something
}
is equivalent to
if(<condition1>) {## do something
} else {## do nothing
}
140.776 Statistical Computing R: Programming
![Page 10: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/10.jpg)
&& (AND) and || (OR) in conditions
Conditions often use && (AND) and || (OR).
if(x>0 && x<1) {y<-x^2
} else {y<-x^4
}
140.776 Statistical Computing R: Programming
![Page 11: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/11.jpg)
&& (AND) and || (OR) in conditions
> x<-c(1>2,2<3,3==4)> x[1] FALSE TRUE FALSE> y<-c(1<2,2<3,3!=4)> y[1] TRUE TRUE TRUE
> x&&y
140.776 Statistical Computing R: Programming
![Page 12: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/12.jpg)
&& (AND) and || (OR) in conditions
> x[1] FALSE TRUE FALSE> y[1] TRUE TRUE TRUE
> x&&y[1] FALSE
> x&y
140.776 Statistical Computing R: Programming
![Page 13: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/13.jpg)
&& (AND) and || (OR) in conditions
> x[1] FALSE TRUE FALSE> y[1] TRUE TRUE TRUE
> x&&y[1] FALSE
> x&y[1] FALSE TRUE FALSE
140.776 Statistical Computing R: Programming
![Page 14: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/14.jpg)
&& (AND) and || (OR) in conditions
&& and || are different from & and |:The shorter form (& and |) performs elementwise comparisonsin much the same way as arithmetic operators.
The longer form (&& and ||) evaluates left to right, examiningonly the first element of each vector. Evaluation proceeds onlyuntil the result is determined.
140.776 Statistical Computing R: Programming
![Page 15: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/15.jpg)
&& (AND) and || (OR) in conditions
1<2 || 2>3 && 1>2
140.776 Statistical Computing R: Programming
![Page 16: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/16.jpg)
&& (AND) and || (OR) in conditions
Compare the following three expressions:
> 1<2 || 2>3 && 1>2[1] TRUE
> (1<2 || 2>3) && 1>2[1] FALSE
> 1<2 || (2>3 && 1>2)[1] TRUE
Why do you obtain different results?
140.776 Statistical Computing R: Programming
![Page 17: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/17.jpg)
&& (AND) and || (OR) in conditions
In R, operators belong to different precedence groups. && hashigher precedence than ||, therefore && is evaluated first.About precedence of operators:
Use help(Syntax) to learn precedence of operators.
Within an expression, operators of equal precedence areevaluated from left to right.
If you are not sure about which operator is evaluated first, Irecommend you to explicitly specify the priority by using ().
There are substantial precedence differences between R and S.For example, in S, &, &&, | and || have equal precedence.
140.776 Statistical Computing R: Programming
![Page 18: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/18.jpg)
Repetitive execution: for loops
for(var in seq) {expr
}
For loops are commonly used for iterating over the element of an object(list, vector, etc.). For example:
for(i in 1:10) {print(i)
}
140.776 Statistical Computing R: Programming
![Page 19: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/19.jpg)
Repetitive execution: for loops
These loops have the same behavior:
x<-c("a","b","c","d")
for(i in 1:4) {print(x[i])
}
for(i in seq_along(x)) {print(x[i])
}
for(letter in x) {print(letter)
}
for(i in 1:4) print(x[i])
140.776 Statistical Computing R: Programming
![Page 20: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/20.jpg)
Find banana
> load("apple-banana-array.rda")
140.776 Statistical Computing R: Programming
![Page 21: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/21.jpg)
Nested loops
Loops can be nested:
x<-matrix(1:60,6,10)
for(i in seq_len(nrow(x))) {for(j in seq_len(ncol(x))) {
print(x[i,j])}
}
140.776 Statistical Computing R: Programming
![Page 22: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/22.jpg)
Repetitive execution: while
while(cond) {expr
}
While loops evaluate a condition repetitively. If the condition is true,then the expression in the loop body is executed. Otherwise, the loop willbe ended. For example:
count<-0while(count<10) {
print(count)count<-count+1
}
140.776 Statistical Computing R: Programming
![Page 23: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/23.jpg)
Repetitive execution: while
Another example:
## simulate a random walkz<-5while(z>=3 && z<=10) {
print(z)coin<-rbinom(1,1,0.5)if(coin == 1) {
z<-z+1} else {
z<-z-1}
}
140.776 Statistical Computing R: Programming
![Page 24: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/24.jpg)
Repetitive execution: repeat
repeat {expr
}
This statement executes the expression in the loop repeatedly until it seesa break. For example:
x0<-1tol<-1e-8repeat {
x1<-computeEstimate()
if(abs(x1-x0)<tol) {break
} else {x0<-x1
}}
140.776 Statistical Computing R: Programming
![Page 25: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/25.jpg)
Breaking a loop: break
The break statement can be used to terminate any loop. It is theonly way to terminate repeat loops. For example:
x0<-1tol<-1e-8err<-10iter<-0while (err>tol) {
x1<-computeEstimate()err<-abs(x1-x0)x0<-x1iter<-iter+1if(iter == 100) {
break}
}
140.776 Statistical Computing R: Programming
![Page 26: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/26.jpg)
next and return
next is used to skip an iteration of a loop.
for(i in 1:5) {if(i<=3) {
next}print(i)
}
[1] 4[1] 5
return signals that a function should exit and return a given value.
140.776 Statistical Computing R: Programming
![Page 27: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/27.jpg)
Find banana
> load("apple-banana-list.rda")
(name="apple", nextnode)|V
(name = "apple", nextnode)|
...V
(name = "banana", nextnode)|
...V
NA
140.776 Statistical Computing R: Programming
![Page 28: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/28.jpg)
Signalling Conditions
There are 4 main functions for signalling or handling conditions(i.e. unusual situations) in R.
message: print a message to the console (not necessarily abad thing)
warning: non-fatal problem; print a message to the console
stop: problem is fatal, execution of the program is halted
try, tryCatch: testing for conditions and executing alternatecode (exception handling)
140.776 Statistical Computing R: Programming
![Page 29: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/29.jpg)
Warnings, Messages
for(i in seq_along(x)) {if(<minor condition>) {
message("a minor condition occurred")}if(<more serious condition>) {
warning("something unusual is going on")}if(<fatal condition>) {
stop("cannot continue, aborting")}
}
140.776 Statistical Computing R: Programming
![Page 30: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/30.jpg)
Good habits
Correct grammarR code: immediately source-able; C code: can be compiledwithout errors
Correct resultsProduce logically correct answer
Code readabilityUse monospace font; <80 characters/line; indent your code;comment your codes
Code efficiencyOrganize into functional modules; keep the code short ifpossible
Computational efficiencyWhoever runs fastest wins
140.776 Statistical Computing R: Programming
![Page 31: R: Programming - Departmentshji/courses/statcomputing/WritingProgram1.pdfControl structures Programming is more than just putting commands you’ve learnt so far into a *.R le. A key](https://reader031.fdocuments.net/reader031/viewer/2022022004/5aaf114b7f8b9a190d8ce293/html5/thumbnails/31.jpg)
Grading criteria
Example:
correctness 60%+ computational efficiency 20%+ readability 10%+ code efficiency 10%= 100%
140.776 Statistical Computing R: Programming