Statistics for osmand.net (2014-11) - refererpages
A new visits is defined as each new incoming visitor (viewing or browsing a page) who was not connected to your site during last 60 mn.
Number of client hosts (IP address) who came to visit the site (and who viewed at least one page).
This data refers to the number of different physical persons who had reached the site.
Number of times a page of the site is viewed (Sum for all visitors for all visits).
This piece of data differs from "hits" in that it counts only HTML pages as oppose to images and other files.
Number of times a page, image, file of the site is viewed or downloaded by someone.
This piece of data is provided as a reference only, since the number of "pages" viewed is often prefered for marketing purposes.
This piece of information refers to the amount of data downloaded by all pages, images and files within your site.
Units are in KB, MB or GB (KiloBytes, MegaBytes or GigaBytes)
Awstats recognizes each access to your site after a search from the 228 most popular Internet Search Engines and Directories (such as Yahoo, Altavista, Lycos, Google, Voila, etc...).
List of all external pages which were used to link (and enter) to your site (Only the 10 most often used external pages are shown).
Links used by the results of the search engines are excluded here because they have already been included on the previous line within this table.
This table shows the list of the most frequent keyphrases or keywords used to find your site from Internet Search Engines and Directories.
(Keywords from the 228 most popular Search Engines and Directories are recognized by Awstats, such as Yahoo, Altavista, Lycos, Google, Voila, etc...).
Note that total number of searches for keywords might be greater than total number of searches for keyphrases (real number of searches) because when 2 keywords were used on same search, search is counted twice for keywords (once for each word).
Robots (sometimes refer to Spiders) are automatic computer visitors used by many search engines that scan your web site to index it and rank it, collect statistics on Internet Web sites and/or see if your site is still online.
Awstats is able to recognize up to 643 robots.
All time related statistics are based on server time.
Here, reported data are: average values (calculated from all data between the first and last visit in analyzed range).
Here, reported data are: cumulative sums (calculated from all data between the first and last visit in analyzed range).
Some Visits durations are 'unknown' because they can't always be calculated. This is the major reason for this:
- Visit was not finished when 'update' occured.
- Visit started the last hour (after 23:00) of the last day of a month (A technical reason prevents Awstats from calculating duration of such sessions)
Worms are automatic computer visitors that are in fact external servers, infected by a virus, that try
to make particular hits on your server to infect it. In most cases, such worms exploit some bugs of not up to date
or commercial servers. If your system is not the sensitive target of the worm, you can simply ignore those hits.
There is very few 'server worms' in the world but they are very active at some times.
Awstats is able to recognize 0 known worm's signatures (nimda,code red,...).
No description for this error.
Request was understood by server but will be processed later.
Server has processed the request but there is no document to send.
Partial content.
Requested document was moved and is now at another address given in answer.
No description for this error.
Syntax error, server didn't understand request.
Tried to reach an URL where a login/password pair was required.A high number within this item could mean that someone (such as a hacker) is attempting to crack, or enter into your site (hoping to enter a secured area by trying different login/password pairs, for instance).
Tried to reach an URL not configured to be reachable, even with an login/password pair (for example, an URL within a directory not defined as "browsable".).
Tried to reach a non existing URL. This error often means that there is an invalid link somewhere in your site or that a visitor mistyped a certain URL.
Server has taken too much time to respond to a request. This error frequently involves either a slow CGI script which the server was required to kill or an extremely congested web server.
Internal error. This error is often caused by a CGI program that had finished abnormally (coredump for example).
Unknown requested action.
Code returned by a HTTP server that works as a proxy or gateway when a real, targeted server doesn't answer successfully to the client's request.
Internal server error.
Gateway Time-out.
HTTP Version Not Supported.
Last Update:&09 Nov 2014 - 21:05
Reported period:
<option value="
<option selected="selected" value="
& Exclude filter&:
Links from an external page (other web sites except search engines) &
Total: 23,909 different pages-urlPagesPercentHitsPercent
7,9159.8 %7,9158.1 %
3,1093.8 %3,1093.2 %
2,5353.1 %7,2797.5 %
1,7732.2 %1,8681.9 %
1,1841.4 %1,1841.2 %
8501 %8500.8 %
8020.9 %5,1455.3 %
7400.9 %7400.7 %
6500.8 %6500.6 %
6280.7 %6280.6 %
5560.6 %5560.5 %
4860.6 %4860.5 %
4640.5 %4640.4 %
4470.5 %4470.4 %
4310.5 %4310.4 %
4300.5 %4300.4 %
2970.3 %2970.3 %
2650.3 %2650.2 %
2510.3 %2510.2 %
2470.3 %2470.2 %
2350.2 %2350.2 %
2180.2 %2180.2 %
2170.2 %2170.2 %
2090.2 %2090.2 %
2070.2 %2070.2 %
2050.2 %2050.2 %
2050.2 %2050.2 %
2040.2 %2040.2 %
2000.2 %2000.2 %
1930.2 %1930.1 %
1930.2 %1930.1 %
1850.2 %1850.1 %
1820.2 %1820.1 %
1770.2 %1770.1 %
1720.2 %1720.1 %
1700.2 %1700.1 %
1640.2 %1640.1 %
1640.2 %1640.1 %
1560.1 %1560.1 %
1370.1 %1370.1 %
1360.1 %1360.1 %
1340.1 %1340.1 %
1340.1 %1340.1 %
1320.1 %2,6782.7 %
1320.1 %1320.1 %
1290.1 %1290.1 %
1270.1 %1270.1 %
1270.1 %7840.8 %
1260.1 %1260.1 %
1260.1 %1260.1 %
1250.1 %1250.1 %
1220.1 %1220.1 %
1220.1 %1220.1 %
1220.1 %1220.1 %
1210.1 %1210.1 %
1210.1 %1210.1 %
1200.1 %1200.1 %
1200.1 %1200.1 %
1200.1 %1200.1 %
1190.1 %1190.1 %
1190.1 %1190.1 %
1190.1 %1190.1 %
1180.1 %1180.1 %
1160.1 %1160.1 %
1140.1 %1140.1 %
1140.1 %1140.1 %
1140.1 %1140.1 %
1140.1 %1140.1 %
1130.1 %1130.1 %
1130.1 %1130.1 %
1130.1 %1130.1 %
1110.1 %1110.1 %
1110.1 %1110.1 %
1110.1 %1110.1 %
1100.1 %1100.1 %
1100.1 %1100.1 %
1100.1 %1100.1 %
1090.1 %1090.1 %
1090.1 %1090.1 %
1080.1 %1080.1 %
1070.1 %1070.1 %
1070.1 %1070.1 %
1060.1 %1060.1 %
1060.1 %1060.1 %
1040.1 %1040.1 %
1030.1 %1030.1 %
1030.1 %1030.1 %
1030.1 %1030.1 %
1020.1 %1020.1 %
1020.1 %1020.1 %
1010.1 %1010.1 %
1010.1 %1010.1 %
1000.1 %1000.1 %
1000.1 %1000.1 %
990.1 %990.1 %
980.1 %980.1 %
980.1 %980.1 %
980.1 %980.1 %
980.1 %980.1 %
910.1 %910 %
890.1 %890 %
890.1 %890 %
880.1 %880 %
860.1 %860 %
860.1 %860 %
840.1 %840 %
820.1 %820 %
820.1 %820 %
800 %800 %
790 %790 %
780 %780 %
770 %770 %
760 %760 %
750 %750 %
710 %710 %
700 %700 %
700 %700 %
690 %690 %
680 %680 %
660 %660 %
650 %650 %
650 %650 %
640 %640 %
600 %600 %
580 %580 %
570 %570 %
570 %660 %
560 %560 %
550 %550 %
550 %550 %
550 %550 %
540 %540 %
540 %540 %
530 %530 %
530 %530 %
530 %530 %
510 %510 %
510 %510 %
510 %510 %
500 %500 %
500 %500 %
500 %500 %
490 %490 %
490 %490 %
480 %480 %
480 %480 %
470 %470 %
470 %470 %
460 %460 %
450 %450 %
450 %450 %
450 %450 %
450 %450 %
450 %450 %
440 %440 %
440 %440 %
440 %440 %
430 %430 %
430 %430 %
420 %420 %
420 %420 %
420 %420 %
420 %420 %
420 %420 %
420 %420 %
420 %420 %
410 %410 %
410 %410 %
410 %410 %
400 %400 %
400 %400 %
400 %400 %
400 %400 %
400 %400 %
400 %400 %
400 %400 %
400 %400 %
400 %400 %
390 %390 %
390 %390 %
390 %390 %
390 %390 %
380 %380 %
380 %380 %
380 %380 %
380 %380 %
370 %370 %
370 %370 %
370 %370 %
370 %370 %
360 %360 %
360 %360 %
360 %360 %
360 %360 %
350 %350 %
350 %350 %
350 %350 %
350 %350 %
350 %350 %
350 %350 %
350 %350 %
350 %350 %
350 %350 %
350 %350 %
350 %350 %
350 %350 %
340 %340 %
340 %340 %
330 %330 %
330 %330 %
330 %330 %
330 %330 %
330 %330 %
330 %330 %
330 %330 %
330 %330 %
320 %320 %
320 %320 %
320 %320 %
320 %320 %
320 %320 %
320 %320 %
310 %310 %
310 %310 %
310 %310 %
300 %300 %
300 %300 %
300 %300 %
300 %300 %
300 %300 %
300 %300 %
300 %300 %
290 %290 %
290 %290 %
290 %290 %
290 %290 %
290 %290 %
290 %290 %
290 %290 %
290 %290 %
290 %290 %
290 %290 %
280 %280 %
280 %280 %
280 %280 %
280 %280 %
280 %280 %
280 %280 %
270 %780 %
270 %270 %
270 %270 %
270 %270 %
270 %270 %
270 %270 %
270 %270 %
260 %260 %
260 %260 %
260 %260 %
260 %260 %
260 %260 %
260 %260 %
260 %260 %
260 %260 %
260 %260 %
260 %260 %
260 %260 %
250 %250 %
250 %250 %
250 %250 %
250 %250 %
240 %240 %
240 %240 %
240 %240 %
240 %240 %
240 %240 %
240 %240 %
240 %240 %
230 %230 %
230 %230 %
230 %230 %
230 %230 %
230 %230 %
230 %230 %
230 %230 %
230 %230 %
230 %230 %
230 %230 %
220 %220 %
220 %220 %
220 %220 %
220 %220 %
220 %220 %
220 %220 %
220 %220 %
220 %220 %
220 %220 %
220 %220 %
220 %220 %
220 %220 %
210 %210 %
210 %210 %
210 %960 %
210 %210 %
210 %210 %
210 %210 %
210 %210 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
200 %200 %
190 %190 %
190 %190 %
190 %190 %
190 %190 %
190 %190 %
190 %190 %
190 %190 %
190 %190 %
190 %190 %
190 %190 %
190 %190 %
190 %190 %
190 %190 %
190 %190 %
190 %190 %
190 %190 %
180 %180 %
180 %180 %
180 %180 %
180 %180 %
180 %180 %
180 %180 %
180 %180 %
180 %180 %
180 %180 %
180 %180 %
180 %3220.3 %
180 %180 %
180 %180 %
180 %180 %
180 %180 %
180 %180 %
180 %180 %
170 %170 %
170 %170 %
170 %170 %
170 %170 %
170 %170 %
170 %170 %
170 %170 %
170 %170 %
170 %170 %
170 %170 %
170 %170 %
170 %170 %
170 %170 %
170 %170 %
170 %170 %
170 %170 %
160 %160 %
160 %160 %
160 %160 %
160 %160 %
160 %160 %
160 %160 %
160 %160 %
160 %160 %
160 %160 %
160 %160 %
160 %160 %
160 %160 %
160 %160 %
160 %160 %
160 %160 %
160 %160 %
160 %160 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %150 %
150 %700 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
140 %140 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %430 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
130 %130 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
120 %120 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
110 %110 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %1160.1 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
100 %100 %
Others30,60738 %33,71034.9 %
Advanced Web Statistics 7.0 (build 1.971) -欢迎来到应用汇！请
Block登陆器Pro
2.5万-2.6万下载 / 33人喜欢 / 37人评论
大小：4.29M 更新：
[我的世界]的第三方登陆器
Block登陆器Pro
BlockLauncher Pro是一款用于Minecraft(MC)的第三方登陆器。这个登陆器允许你加载更多的MOD，皮肤和纹理。使用登陆器前请确保正确安装了MINECRAFT PE。
-无 PTPatches 和 ModPE scripts 数量限制
-能够从任何手机版材质包中加载材质
-修正服务IP
Added support for Minecraft PE 0.9.5 (and removed support for earlier versions)
Fixed setSneaking
Added command history
This version should be able to run many ModPE scripts originally for MCPE 0.8.1 on 0.9.5. Any incompatible scripts should be posted on the forum thread.
14EUR 绝望。尊
为什么我会闪退？
皇族ちミ单身汉
怎么会闪退啊
┻_℡Manぃ囄開゛
快，快，快，逮到楼主了。快拿绳子来，架锅，架锅，多放点水 不要放油，楼主太肥了，艹！ 你脱裤子干嘛？这个不能日！对对对！柴火烧大 ！快都来帮把手把楼主抬锅里去！一 二 三，走你！艹 怎么扔火里了。
梦毁网络、i
怎么使用这个。求介绍。
Dragon-黒皇绝望
赞！！！！！！！非常好！英语渣慎用！
I＇am joe!!!
赞赞赞，超好用
1.7的我有，在百度的“我的世界pe”这个贴吧去搜吧
我一进就停止运行
亲，想发表评论请下载哦～
发现该应用有下载安装使用错误或恶意扣费携带病毒，请
2.4万-2.5万下载
1万-1.1万下载
3.4万-3.5万下载
AppChina 应用汇QQ群：
应用汇开发者交流开发群：
换量合作QQ：
微博联系人QQ：
网游合作QQ：
单机合作QQ：
友情链接QQ：
反馈邮箱：
广告合作：Some Useful Scripts
Time Series Analysis and Its Applications: With R Examples
Second Edition
For a similar
plot with one time series, use lag.plot1. The full call is lag.plot1(x,m,corr=TRUE,smooth=TRUE)
and it will generate a grid of scatterplots of
x(t-h) versus x(t) for h = 1,...,m, along with
the autocorrelation values in blue and a lowess fit in red.
The defaults are the same as in the previous script.
If you don't want any
correlations or lines, simply use R's lag.plot.
Here's an example
assuming soi is still available.
lag.plot1(soi,12,smooth=TRUE)
which results in the graph:
Note that lag.plot1(soi,12) will give you the graphic without the smoothed (red) lines.
Now, here's an example using acf2 on the Recruitment series (rec):
# computes to lag sqrt(n)+10
0.78 -0.44
0.63 -0.05
[31,] -0.13
[32,] -0.11
you get the values as above ... and a picture:
You can specify the number of lags... 48 for example:
acf2(rec,48)
0.78 -0.44
0.63 -0.05
0.20 -0.04
and the picture:
sarima & sarima.for
The basics are as follows:
If your time series is in x and you want to fit an ARIMA(p,d,q) model to your data, the call is
sarima(x,p,d,q).
The results are the parameter estimates, standard errors, AIC, AICc, BIC
(as defined in Chapter 2) and diagnostics.
To fit a seasonal ARIMA model, the call is sarima(x,p,d,q,P,D,Q,s).
example, sarima(x,2,1,0) will fit an ARIMA(2,1,0) model to the series in x, and
sarima(x,2,1,0,0,1,1,12) will fit a seasonal
ARIMA(2,1,0)&(0,1,1)12 model
to the series in x.
Here are some examples using sarima and sarima.for.
First, suppose I
generate the random walk with drift, x, mentioned in :
set.seed(1)
# so you can reproduce the results
v = rnorm(100,1,1)
# v contains 100 iid N(1,1) variates
x = cumsum(v)
# x is a random walk with drift = 1
and I try to fit an ARIMA(1,1,0) model:
sarima(x,1,1,0)
# <-- this is the most basic form ... easy, huh?
value -0.105745
<-- sarima shows the output from the optimization
2 value -0.105750
<-- this is the Conditional Sum Sq (CSS) part
3 value -0.105750
** you can shut this off using sarima(x,1,1,0,details=FALSE)
4 value -0.105750
... additional
are at the end of these examples
5 value -0.105750
6 value -0.105751
7 value -0.105751
7 value -0.105751
7 value -0.105751
value -0.105751
value -0.110799
<-- this is the Max Likelihood (ML) part
1 value -0.110799
value -0.110799
arima(x = data, order = c(p, d, q), seasonal = list(order = c(P, D, Q), period = S),
xreg = constant, include.mean = F, optim.control = list(trace = 1, REPORT = 1, reltol=tol))
Coefficients:
<-- notice this says constant ()
<-- and a constant is fit even when
differencing ()
sigma^2 estimated as 0.8012:
log likelihood = -129.51,
aic = 265.01
[1] 0.8184023
<-- this is AIC as defined in Chapter 2
(type ?AIC to see how R computes aic - the two are comparable)
[1] 0.8409023
<-- this is AICc as defined in Chapter 2
[1] -0.1294943
<-- this is SIC or BIC as defined in Chapter 2
The fit is like
&x(t) = 1.1169 -.0031*&x(t-1) + w(t)
with the estimate of
&2 being .8012
and the AR term is not significant, as expected.
Note that the command will also generate diagnostics as a graphic, like this:
This graphic is an improved version of R's tsdiag() script.
In addition to adding a normal Q-Q plot of the standardized residuals,
the correct degrees of freedom for calculating the p-values ( for more details).
If you want to play with the innovations (i.e., the residuals) from the fit, they're
stored in innov:
Time Series:
Frequency = 1
0. -0. -2.
[16] -0. -0.
0. -0. -2.
[26] -0. -0. -1. -0.
0. -0. -1.
[36] -0. -0. -0.
[41] -0. -0.
1. -0. -1.
0. -0. -0.
0. -1. -1.
0. -0. -0. -0. -0.
[81] -0. -0.
0. -1. -0. -1. -0.
Another way to get the innovations is to give the analysis a name, e.g.,
dog=sarima(x,1,1,0), and then you can pull out the innovations
as dog$fit$resid or resid(dog$fit) or even
resid(sarima(x,1,1,0)$fit) if you want to leave out the dog.
Once I'm satisfied with a particular model, I can forecast. For example,
with the model above and n.ahead=5 I get (this will also produce a plot
of data, the forecasts and corresponding error bounds):
sarima.for(x,5,1,1,0)
Time Series:
Start = 101
Frequency = 1
[1] 112.4 114.2 116.4751
Time Series:
Start = 101
Frequency = 1
[1] 0.....9966174
... and the graphic:
Next, let's fit an AR(2) to the Recruitment data (minus all of the graphics):
sarima(rec,2,0,0)
value 3.332380
2 value 3.251366
3 value 2.564654
4 value 2.430141
5 value 2.258212
6 value 2.253343
7 value 2.248346
8 value 2.248345
9 value 2.248345
10 value 2.248341
11 value 2.248332
12 value 2.248331
13 value 2.248330
13 value 2.248330
13 value 2.248330
CSS converged in 13 (or 14 or 15 ???) steps
value 2.248330
value 2.248862
2 value 2.248857
3 value 2.248855
4 value 2.248855
5 value 2.248854
6 value 2.248854
7 value 2.248854
8 value 2.248854
9 value 2.248853
10 value 2.248853
10 value 2.248853
10 value 2.248853
and from there, ML converged in 10-12 steps
value 2.248853
arima(x = data, order = c(p, d, q), seasonal = list(order = c(P, D, Q), period = S),
xreg = xmean, include.mean = F, optim.control = list(trace = 1, REPORT = 1, reltol=tol))
Coefficients:
sigma^2 estimated as 89.33:
log likelihood = -1661.51,
aic = 3331.02
[1] 5.505631
[1] 5.510243
[1] 4.532889
The fitted model is like
x(t) = 61..3512-(-.4612)) + 1.3512*x(t-1) - .4612*x(t-2) + w(t)
Sorry, I didn't feel like trying to get sarima to report the constant here.
At least the wording is correct ().
And again, if you would rather not see the output from the nonlinear optimization routine, you would use
sarima(rec,2,0,0,details=FALSE).
Now forecast 5 ahead:
sarima.for(rec,5,2,0,0)
Time Series:
Start = 454
Frequency = 1
[1] 20.36 32.74 44.30006
Time Series:
Start = 454
Frequency = 1
9.....393693
Because there are more than 100 observations, for your viewing pleasure,
the graphic will show only the
final 100 observations and the forecasts (and error bounds):
Here's a seasonal model fit on the recruitment data, minus the diagnostic graphics and with the details off (this is just an
example, it's not a very good model).
sarima(rec,2,0,0,1,0,0,12, details=FALSE)
#<-- it's an ARIMA(2,0,0)&(1,0,0)12
arima(x = xdata, order = c(p, d, q), seasonal = list(order = c(P, D, Q), period = S),
xreg = xmean, include.mean = F, optim.control = list(trace = trc, REPORT = 1,
reltol = tol))
Coefficients:
sigma^2 estimated as 87.86:
log likelihood = -1657.85,
aic = 3325.71
[1] 5.493421
[1] 5.498132
[1] 4.529764
& In sarima,
if there's no differencing (d=0 and D=0) you get the mean estimate.
If there's differencing of order one (either d=1 or D=1, but not both), you'll get the constant term.
In any other situation, no constant or mean term is included in the model.
The idea is that if
you difference more than once (d + D > 1), any drift would have been removed.
& The relative tolerance (reltol) used to assess convergence
in sarima.R and sarima.for.R
is set to sqrt(.Machine$double.eps), the R default.
For details, see the help file for optim under the control
arguments.
You can change this without changing the code.
For example, sarima(rec,2,0,0,tol=.0001) will
do the second example with reltol = .0001.
If there are many parameters
to estimate (e.g., seasonal models), the analysis may take a long time
using the default.
That's not a problem if you're sitting at your desk, but if you're doing a demonstration you
may need a good joke to tell while you're waiting for convergence. The one I thought of on the spot was: This is
why cigarettes were invented.
They last about as long as it takes to get results.
The joke went over pretty
well- you can use it free of charge.
And finally, here are some examples using spec.arma, which will produce the spectral density of an
ARMA model.
It can also be used to determine if the model is causal and/or invertible (and there's a check for common zeros). The basic call is spec.arma(ar, ma) where
ar and ma are vectors containing the model parameters. For example, to see the logged spectral density
of the ARMA(2,1) model x(t) = 1.4 x(t-1) - .9 x(t-2) + w(t) + .8 w(t-1), you issue the command:
spec.arma(ar=c(1.4,-.9), ma=.8)
and you would see
If you don't want the logged spectrum, use spec.arma(ar=c(1.4,-.9), ma=.8, log="no").
The script will test the roots and you'll get warnings if the roots are not outside the unit circle
or if there is parameter redundancy.
For example,
spec.arma(ar=1)
WARNING: Model Not Causal
Error in spec.arma(ar = 1) : Try Again
spec.arma(ma=1)
WARNING: Model Not Invertible
Error in spec.arma(ma = 1) : Try Again
spec.arma(ar=39.95, ma=c(.9,2))
WARNING: Model Not Causal
WARNING: Model Not Invertible
Error in spec.arma(ar = 39.95, ma = c(0.9, 2)) : Try Again
# common zeros:
spec.arma(ar=.8, ma=-.8) # this will work but you'll see the spectral density of white noise
WARNING: Parameter Redundancy
... and you'll get a warning.
# approximate parameter redundancy (common zeros):
spec.arma(ar=.8, ma=-.7999)
WARNING: Parameter Redundancy
note that the spectral ordinates are between
0.9999 and 1.0010, so you basically have white noise
# no arguments gives the spectral density of white noise with &sigma = 1;
spec.arma()
# and if you want to change the variance, do something like this:
spec.arma(ar = c(1.5,-.75), var.noise = 39.95, log="no")
Finally, the frequencies and the spectral density ordinates are returned invisibly, e.g.,
spec.arma(ar=.9)$freq and spec.arma(ar=.9)$spec, if you're interested in
the actual values.