Canonical Correlation and Its Relationship to Discriminant Analysis and Multiple Regression

To determine the predictive potential of selected cognitive and psychomotor variables to estimate basketball performance, 50 male high school players from 3 schools in the same conference were evaluated. One team won the Iowa state championship; the second team had a 12 and 10 record while the third team had a 4 and 16 record. The 3 coaches rated each player's ability from 1 to 10. Multiple regression analysis to predict coaches' rating of ability from vertical jump, hand reaction time, weight, and playing experience gave an R of .76. However, discriminant analysis to classify players on the 3 teams indicated as important knowledge about basketball, dribbling, shooting accuracy, and height. The canonical correlation for the 4 variables and team membership was .64. Using the 4 variables, 60% of the players could be correctly classified to their teams.

Download Full-text

Parental representation in twins discordant for schizophrenia

Psychological Medicine ◽

10.1017/s0033291700028439 ◽

1993 ◽

Vol 23 (2) ◽

pp. 335-340 ◽

Cited By ~ 18

Author(s):

Sidsel Onstad ◽

Ingunn Skre ◽

Svenn Torgersen ◽

Einar Kringlen

Keyword(s):

Regression Analysis ◽

Discriminant Analysis ◽

Multiple Regression Analysis ◽

Multiple Regression ◽

Parental Bonding ◽

Important Variable ◽

Stepwise Discriminant Analysis ◽

Parental Bonding Instrument ◽

Parental Representation

SynopsisParental representation was assessed with the Parental Bonding Instrument (PBI) in 12 monozygotic (MZ) and 19 dizygotic (DZ) same-sexed twin pairs discordant for DSM-III-R schizophrenia. The schizophrenic twins reported less care and more overprotection from both parents than the non-schizophrenic co-twins. Multiple regression analysis disclosed that the results were independent of sex and age. Furthermore, the analysis demonstrated that whether the twins were monozygotic or dizygotic had no impact on the results. A stepwise discriminant analysis showed that difference in perceived paternal protection was the most important variable distinguishing between schizophrenic probands and their non-schizophrenic co-twins.

Download Full-text

Individual differences in cerebral metabolic patterns during pharmacotherapy in obsessive-compulsive disorder: A multiple regression/discriminant analysis of positron emission tomographic data

Biological Psychiatry ◽

10.1016/0006-3223(93)90069-p ◽

1993 ◽

Vol 34 (11) ◽

pp. 798-809 ◽

Cited By ~ 15

Author(s):

N.P. Azari ◽

P. Pietrini ◽

B. Horwitz ◽

K.D. Pettigrew ◽

H.L. Leonard ◽

...

Keyword(s):

Individual Differences ◽

Discriminant Analysis ◽

Multiple Regression ◽

Obsessive Compulsive Disorder ◽

Positron Emission Tomographic ◽

Obsessive Compulsive ◽

Compulsive Disorder ◽

Positron Emission ◽

Tomographic Data

Download Full-text

Local Discriminant Hyperalignment for multi-subject fMRI data alignment

10.1101/092247 ◽

2016 ◽

Cited By ~ 2

Author(s):

Muhammad Yousefnezhad ◽

Daoqiang Zhang

Keyword(s):

Discriminant Analysis ◽

Canonical Correlation ◽

State Of The Art ◽

Experimental Studies ◽

Superior Performance ◽

Fmri Data ◽

Data Alignment ◽

Cognitive States ◽

Multivariate Pattern ◽

The Brain

AbstractMultivariate Pattern (MVP) classification can map different cognitive states to the brain tasks. One of the main challenges in MVP analysis is validating the generated results across subjects. However, analyzing multi-subject fMRI data requires accurate functional alignments between neuronal activities of different subjects, which can rapidly increase the performance and robustness of the final results. Hyperalignment (HA) is one of the most effective functional alignment methods, which can be mathematically formulated by the Canonical Correlation Analysis (CCA) methods. Since HA mostly uses the unsupervised CCA techniques, its solution may not be optimized for MVP analysis. By incorporating the idea of Local Discriminant Analysis (LDA) into CCA, this paper proposes Local Discriminant Hyperalignment (LDHA) as a novel supervised HA method, which can provide better functional alignment for MVP analysis. Indeed, the locality is defined based on the stimuli categories in the train-set, where the correlation between all stimuli in the same category will be maximized and the correlation between distinct categories of stimuli approaches to near zero. Experimental studies on multi-subject MVP analysis confirm that the LDHA method achieves superior performance to other state-of-the-art HA algorithms.

Download Full-text