thr3ads.net - R help - [R] Request for functions to calculate correlated factors influencing an outcome. [May 2015]

If this information is useful, please help other people find it:
Share via:

Lalitha Viswanathan

2015-May-03 09:19 UTC

[R] Request for functions to calculate correlated factors influencing an outcome.

Hi
I have a dataset of the type attached.
Here's my code thus far.
dataset <-data.frame(read.delim("data", sep="\t",
header=TRUE));
newData<-subset(dataset, select = c(Price, Reliability, Mileage, Weight,
Disp, HP));
cor(newData, method="pearson");
Results are
                 Price Reliability    Mileage     Weight       Disp
HP
Price        1.0000000          NA -0.6537541  0.7017999  0.4856769
 0.6536433
Reliability         NA           1         NA         NA         NA
NA
Mileage     -0.6537541          NA  1.0000000 -0.8478541 -0.6931928
-0.6667146
Weight       0.7017999          NA -0.8478541  1.0000000  0.8032804
 0.7629322
Disp         0.4856769          NA -0.6931928  0.8032804  1.0000000
 0.8181881
HP           0.6536433          NA -0.6667146  0.7629322  0.8181881
 1.0000000

It appears that Wt and Price, Wt and Disp, Wt and HP, Disp and HP, HP and
Price are strongly correlated.
To find the statistical significance,
I am trying  sample.correln<-cor.test(newData$Disp, newData$HP,
method="kendall", exact=NULL)
Kendall's rank correlation tau

data:  newx$Disp and newx$HP
z = 7.2192, p-value = 5.229e-13
alternative hypothesis: true tau is not equal to 0
sample estimates:
      tau
0.6563871

If I try the same with
sample.correln<-cor.test(newData$Disp, newData$HP,
method="pearson",
exact=NULL)
I get Warning message:
In cor.test.default(newx$Disp, newx$HP, method = "spearman", exact =
NULL) :
  Cannot compute exact p-value with ties> sample.correln
Spearman's rank correlation rho

data:  newx$Disp and newx$HP
S = 5716.8, p-value < 2.2e-16
alternative hypothesis: true rho is not equal to 0
sample estimates:
      rho
0.8411566

I am not sure how to interpret these values.
Basically, I am trying to figure out which combination of factors
influences efficiency.

Thanks
Lalitha
-------------- next part --------------
Price	Country	Reliability	Mileage	Type	Weight	Disp.	HP


8895	USA	4	33	Small	2560	97	113


7402	USA	2	33	Small	2345	114	90


6319	Korea	4	37	Small	1845	81	63


6635	Japan/USA	5	32	Small	2260	91	92


6599	Japan	5	32	Small	2440	113	103


8672	Mexico	4	26	Small	2285	97	82


7399	Japan/USA	5	33	Small	2275	97	90


7254	Korea	1	28	Small	2350	98	74


9599	Japan	5	25	Small	2295	109	90


5866	Japan	NA	34	Small	1900	73	73


8748	Japan/USA	5	29	Small	2390	97	102


6488	Japan	5	35	Small	2075	89	78


9995	Germany	3	26	Small	2330	109	100


11545	USA	1	20	Sporty	3320	305	170


9745	USA	1	27	Sporty	2885	153	100


12164	USA	1	19	Sporty	3310	302	225


11470	USA	3	30	Sporty	2695	133	110


9410	Japan	5	33	Sporty	2170	97	108


13945	Japan	5	27	Sporty	2710	125	140


13249	Japan	3	24	Sporty	2775	146	140


10855	USA	NA	26	Sporty	2840	107	92


13071	Japan	NA	28	Sporty	2485	109	97


18900	Germany	NA	27	Compact	2670	121	108


10565	USA	2	23	Compact	2640	151	110


10320	USA	1	26	Compact	2655	133	95


10945	USA	4	25	Compact	3065	181	141


9483	USA	2	24	Compact	2750	141	98


12145	Japan/USA	5	26	Compact	2920	132	125


12459	Japan/USA	4	24	Compact	2780	133	110


10989	Japan	5	25	Compact	2745	122	102


17879	Japan	4	21	Compact	3110	181	142


11650	Japan	5	21	Compact	2920	146	138


9995	USA	2	23	Compact	2645	151	110


15930	France	NA	24	Compact	2575	116	120


11499	Japan/USA	5	23	Compact	2935	135	130


11588	Japan/USA	5	27	Compact	2920	122	115


18450	Sweden	3	23	Compact	2985	141	114


24760	Japan	5	20	Medium	3265	163	160


13150	USA	3	21	Medium	2880	151	110


12495	USA	2	22	Medium	2975	153	150


16342	USA	3	22	Medium	3450	202	147


15350	USA	2	22	Medium	3145	180	150


13195	USA	3	22	Medium	3190	182	140


14980	USA	1	23	Medium	3610	232	140


9999	Korea	NA	23	Medium	2885	143	110


23300	Japan	5	21	Medium	3480	180	158


17899	Japan	5	22	Medium	3200	180	160


13150	USA	2	21	Medium	2765	151	110


14495	USA	NA	21	Medium	3220	189	135


21498	Japan	3	23	Medium	3480	180	190


16145	USA	3	23	Large	3325	231	165


14525	USA	1	18	Large	3855	305	170


17257	USA	3	20	Large	3850	302	150


13995	USA	NA	18	Van	3195	151	110


15395	USA	3	18	Van	3735	202	150


12267	USA	3	18	Van	3665	182	145


14944	Japan	5	19	Van	3735	181	150


14929	Japan	NA	20	Van	3415	143	107


13949	Japan	NA	20	Van	3185	146	138


14799	Japan	NA	19	Van	3690	146	106

Michael Dewey

2015-May-03 11:24 UTC

head link

[R] Request for functions to calculate correlated factors influencing an outcome.

Dear Lalitha, see inline below

On 03/05/2015 10:19, Lalitha Viswanathan wrote:> Hi
> I have a dataset of the type attached.
> Here's my code thus far.
> dataset <-data.frame(read.delim("data", sep="\t",
header=TRUE));
> newData<-subset(dataset, select = c(Price, Reliability, Mileage, Weight,
> Disp, HP));
In fact in the file the variable seems to be called Disp.
> cor(newData, method="pearson");
> Results are
>                   Price Reliability    Mileage     Weight       Disp
> HP
> Price        1.0000000          NA -0.6537541  0.7017999  0.4856769
>   0.6536433
> Reliability         NA           1         NA         NA         NA
> NA
> Mileage     -0.6537541          NA  1.0000000 -0.8478541 -0.6931928
> -0.6667146
> Weight       0.7017999          NA -0.8478541  1.0000000  0.8032804
>   0.7629322
> Disp         0.4856769          NA -0.6931928  0.8032804  1.0000000
>   0.8181881
> HP           0.6536433          NA -0.6667146  0.7629322  0.8181881
>   1.0000000
>
> It appears that Wt and Price, Wt and Disp, Wt and HP, Disp and HP, HP and
> Price are strongly correlated.
> To find the statistical significance,
> I am trying  sample.correln<-cor.test(newData$Disp, newData$HP,
> method="kendall", exact=NULL)
> Kendall's rank correlation tau
>
> data:  newx$Disp and newx$HP
> z = 7.2192, p-value = 5.229e-13
> alternative hypothesis: true tau is not equal to 0
> sample estimates:
>        tau
> 0.6563871
>
> If I try the same with
> sample.correln<-cor.test(newData$Disp, newData$HP,
method="pearson",
> exact=NULL)
When I try that it works fine.
The real question is why when you asked it for the Pearson coefficient 
it decided to give you the Spearman as the warning message below points 
out. I suspect you have done something else which you did not tell us about.
> I get Warning message:
> In cor.test.default(newx$Disp, newx$HP, method = "spearman",
exact = NULL) :
>    Cannot compute exact p-value with ties
>> sample.correln
>
> Spearman's rank correlation rho
>
> data:  newx$Disp and newx$HP
> S = 5716.8, p-value < 2.2e-16
> alternative hypothesis: true rho is not equal to 0
> sample estimates:
>        rho
> 0.8411566
>
> I am not sure how to interpret these values.
> Basically, I am trying to figure out which combination of factors
> influences efficiency.
>
> Thanks
> Lalitha
>
>
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
-- 
Michael
http://www.dewey.myzen.co.uk/home.html

R help - May 2015 - Request for functions to calculate correlated factors influencing an outcome.

[R] Request for functions to calculate correlated factors influencing an outcome.

[R] Request for functions to calculate correlated factors influencing an outcome.