You Can Call Me Now : Basic&Applied Stat

purpler

Originally posted by DDT_pacak85 at 19-11-2006 11:43 PM
camner plak kes yg ader pasal critical value ngan accepted value .. yg kener pakai outlier utk kiter tau sama ada nilai tu kiter accepted or rejected .. susah la ... ajar skit

ni guna test statistics apa ek???
Biasanya kita bole dapatkan critical value daripada table test statistics yg kita gunakan sama ada guna t-test, chi-square test dn sebagainya.
kalau basic...macamnilah.....

is known as the critical value, the positive z value that is at the vertical boundary for the area of in the right tail of the standard normal distribution.






kalau buat hyphothesis testing(ni salah satu inferential statistics)....

• Kenalpasti null hypothesis dan alternative hypothesis
  
  
  
• Kenalpasti level of significance,. Biasanya, kita guna 0.01 (1% strict), 0.05(5%-moderate) or 0.10 (10%-not strict).
  
  
  
• compute the test statistics from the data available.
  
  
• Kenal pasti p-value dan bandingkan dgn .p-value ialah kebarangkalian kita akan rejecting a true null berdasarkan test statistics yg kita gunakan.
  
                                                     Decision
                            Do not accept H0                  Accept H0
           H0             Type I Error                   Right Decision
            H1            Right Decision                Type II Error
  
  H0=null is true
  H1=null is false
  Type I error=occurs when the null hypothesis is not accepted when it is in fact true.
  Type II error=occurs when the null hypothesis H0 is accepted when it is in fact false.
  
  
• Kenalpasti adakah cukup syarat kita bole reject(not accepted lebih elok digunakan) the null hypothesis. Sebaik²nya result bole reject the null hypothesis. Kalau tak bole reject pun takpe.
  
  
• buat kesimpulan berdasarkan apa yg data kita nak.
  

[ Last edited by  purpler at 1-12-2006 12:55 PM ]

purpler

sape² ada masalah buat SAS atau SPSS....
bolelah tanya aku....
InsyaAllah aku bole tolong...

Sinaran

Purpler : your info is very good. tq.

As a decision maker....banyak perkara yang kita perlu fikirkan. kadang-kadang ada ianya dalam keadaan pasti (certainty), kadang-kadang dalam keadaan tak pasti (uncertainty).. betulkan?So...did u know bahawa Statistics is the Measurement of Uncertainty?

Come back to your problem..since that, u have to find out the 'proof', then..we need to do a research? it is?

Jika ianya berkaitan our account...just a open the operating cost, balance sheet dan sekutu2nya, jika ianya berkenaan price of product, just look at your info about supply and the demand.

Tapi..jika ia berkenaan (1) customer satisfaction index,(2) keberkesanaan kaedah pengajaran selepas menghadiri extra class, (3) penentuan samada terdapat perbezaan harga product A bagi tiga supermarket, (4) menentukan faktor yang mempengaruhi harga getah dan lain-lain, (5) mengenal pasti sama ada jalan bersimpang merupakan penyumbang kepada peningkatan kadar kemalangan....apa yg kita patut lakukan?

(bersambung...)

[ Last edited by  Sinaran at 2-12-2006 11:28 AM ]

Sinaran

Sambungan:

untuk kes-kes itu. kita pun kena identify the population and sample of study(recall samplin techniques). then..will involve dengan data collection (how to collect?..kalau ada input..i'll share it with u), data preparation and screening (key in in the statistical package...yg ni..minta tolong purpler..nanti i tolong kalau mampu).


dah selesai..maka..bermulalah data analysis. jeng jeng jeng..masa inilah kita kena pastikan univariate (satu variable), bivariate(dua variable) dan multivarite(lebih atau sama dgn dua variable).

then..kena tahu juga skala pengukuran bagi data untuk sesuatu pemboleh ubah sama ada nominal, ordinal, interval atau ratio...kenapa? sebab setiap analysis dalam statistik ada assumption. bukan semua keadaan boleh guna t-test, bukan semua keadaan boleh guna ANOVA, regression dsb (ini kena berguru dgn saya...tak mahal..satu perkataan RM100 je..hhehhehhe)


sebelum teruskan...tahukah anda?
1) Parametric test perlukan normallity assumption
2) Non parametric test adalah untuk free distribution

Oleh itu, kita perlu "normallity test (Shapiro Wilk test/kolmogorov Smirnov test etc)

kemudian pilihla parameteric or non parametric test.

psttt..jika guna SPSS/Sas/minitab etc..jgn main klik je. tanpa fikirkan assumption untuk pengujiannya. pakej akan menjanakan output walau dalam apa keadaan sekalipun.

sambung...

[ Last edited by  Sinaran at 2-12-2006 11:44 AM ]

Sinaran

Sambungan:

seterusnya..kita teruskan dengan analysis...pada masa itu..kita pun akan terfikir, nak analysis apa ye? fikir dan fikir...rujuk kepada objektif kajian..kebiasaanya (ayat mudah la ye) objektif adalah terdiri drpd senarai pertanyaan2  yg nak diketahui.

pertanyaan-pertanyaan kita itulah..yg dikatakan hypothesis null (ayat mudah >>hypotesis ckp kosong: yg mungkin betul, mungkin tidak betul)

based on apa yg nak diketahui...dan method (statistical testing : parametric &nonparametric tes) yg sesuai dengan nya...maka kita pun run the programme, utk generate the output). dunia hari ini...computer output akan memberikan maklumat mengenai p-value (probability value). untuk SAS rujuk pada Prob-value, untuk SPSS rujuk pada Sig. value. Anyway, apa itu p-value???
bersambung...

Sinaran

apa itu p-value?

P value is associated with a test statistic. It is "the probability, if the test statistic really were distributed as it would be under the null hypothesis, of observing a test statistic [as extreme as, or more extreme than] the one actually observed." The smaller the P value, the more strongly the test rejects the null hypothesis, that is, the hypothesis being tested.

A p-value of .05 or less rejects the null hypothesis "at the 5% level" that is, the statistical assumptions used imply that only 5% of the time would the supposed statistical process produce a finding this extreme if the null hypothesis were true.5% and 10% are common significance levels to which p-values are compared

What is "statistical significance" (p-value). The statistical significance of a result is the probability that the observed relationship (e.g., between variables) or a difference (e.g., between means) in a sample occurred by pure chance ("luck of the draw"), and that in the population from which the sample was drawn, no such relationship or differences exist. Using less technical terms, one could say that the statistical significance of a result tells us something about the degree to which the result is "true" (in the sense of being "representative of the population"). More technically, the value of the p-value represents a decreasing index of the reliability of a result (see Brownlee, 1960). The higher the p-value, the less we can believe that the observed relation between variables in the sample is a reliable indicator of the relation between the respective variables in the population. Specifically, the p-value represents the probability of error that is involved in accepting our observed result as valid, that is, as "representative of the population." For example, a p-value of .05 (i.e.,1/20) indicates that there is a 5% probability that the relation between the variables found in our sample is a "fluke."

http://www.statsoft.com/textbook/esc.html

purpler

sinaran ni student statistics ke???:hmm:

Sinaran

Originally posted by purpler at 2-12-2006 03:45 PM
sinaran ni student statistics ke???:hmm:

aha...graduate from UItm

purpler

Originally posted by Sinaran at 4-12-2006 02:12 PM



aha...graduate from UItm

diploma ke degree???
tahun bila grad????:nerd:

sori mod...off topic jap....

ff:

Sinaran

Originally posted by purpler at 4-12-2006 04:03 PM



diploma ke degree???
tahun bila grad????:nerd:

sori mod...off topic jap....

ff:

ahakss....jejak kasih ke? rahsia la dulu.
hint: masa i graduate, school engin (opposite ftmsk) belum dibina.

Sinaran

FOUR TYPES OF MEASUREMENT SCALES ARE: nominal, ordinal, interval, and ratio (NOIR). It's important to know which type of scale is represented in the research data because different scales require different methods of statistical analysis. All variables are represented in at least one of the above measurement scale. That is, the variables represented nominal scale are named nominal variables, the variables represented ordinal scale are called ordinal variables and so forth.

Nominal Variables

Nominal variables, also called categorical variables, represent the lowest level of measurement. They simply classify persons or objects into two or more categories where members of a category have at least one common characteristic. Nominal variables include gender (female, male); employment status (full-time, part-time, unemployed); marital status (married, divorced, single); and type of school (public, private, charter). For identification purposes, nominal variables are often represented by numbers. For example, the category "male" may be represented by number 1 and "female" by the number 2. It is critically important to understand that such numbering of nominal variables does not indicate that one category is higher or better than another. The numbers are only labels for the groups.

Ordinal Variables

Ordinal variables, like nominal variables, classify persons or objects but also rank them in terms of the degree to which they possess a characteristic of interest.In other words, ordinal variables put persons or objects in order from highest to lowest or from most to least.

However, ordinal variables do not indicate how much higher or how much better one person performed compared to another. In other words, intervals between ranks are not equal; the difference between rank 1 and rank 2 is not necessarily the same as the difference between rank 2 and rank 3.

References:
1. http://coe.sdsu.edu/eet/articles/measurescales/index.htm
2. http://www.csse.monash.edu.au/~smarkham/resources/scaling.htm
3. http://www.math.sfu.ca/~cschwarz/Stat-301/Handouts/node5.html

Sinaran

sorry..silap enter

[ Last edited by  Sinaran at 6-12-2006 01:37 PM ]

purpler

Originally posted by Sinaran at 4-12-2006 04:43 PM


ahakss....jejak kasih ke? rahsia la dulu.
hint: masa i graduate, school engin (opposite ftmsk) belum dibina.

masa purpler grad....
skul FTMSK yg baru sedang dibina....
kiranya ni seniorla...hehe!!!:nerd:

Sinaran

Structural Equation Modeling (SEM) is a multivariate procedure that, as defined by Ullman (1996), 揳llows examination of a set of relationships between one or more independent variables, either continuous or discrete, and one or more dependents variables, either continuous or discrete.