24
Thresholds and ordinal data Sarah Medland – Boulder 2010

Thresholds and ordinal data

  • Upload
    libra

  • View
    35

  • Download
    0

Embed Size (px)

DESCRIPTION

Thresholds and ordinal data. Sarah Medland – Boulder 2010. A plug for OpenMx?. Very few packages can handle ordinal data adequately… OpenMx can also be used for more than just genetic analyses Regression Polycohoric correlations Factor analysis. Binary data. File: two_cat.dat - PowerPoint PPT Presentation

Citation preview

Page 1: Thresholds and ordinal data

Thresholds and ordinal data

Sarah Medland – Boulder 2010

Page 2: Thresholds and ordinal data

A plug for OpenMx?

Very few packages can handle ordinal data adequately…

OpenMx can also be used for more than just genetic analyses Regression Polycohoric correlations Factor analysis...

Page 3: Thresholds and ordinal data

Binary data

File: two_cat.dat Zyg twin1 twin2 Age Sex

Trait – lifetime cannabis use Never Smoked/Ever Smoked

Page 4: Thresholds and ordinal data

Twin 1 cannabis use

0 = never usedtwin1

822 47.5 53.0 53.0

730 42.2 47.0 100.0

1552 89.7 100.0

179 10.3

1731 100.0

0

1

Total

Valid

SystemMissing

Total

Frequency Percent Valid PercentCumulative

Percent

Page 5: Thresholds and ordinal data

Twin 1 cannabis use

Page 6: Thresholds and ordinal data

Twin 1 cannabis use

Liability or ‘risk’ of initiation distribution

Just because an individual has never used cannabis does not mean their ‘risk’ of initiation is zero

Page 7: Thresholds and ordinal data

Mean = .47

SD =.499

Non Smokers =53%

The observed phenotype is an imperfect measurement of an underlying continuous distribution

ie Obesity vs BMIMDD vs quantitative depression scales

Page 8: Thresholds and ordinal data

Raw data distribution

Mean = .47

SD =.499

Non Smokers =53%

Threshold =.53

Standard normal distribution

Mean = 0

SD =1

Non Smokers =53%

Threshold =.074

Page 9: Thresholds and ordinal data

Threshold = .074 – Huh what?

How can I work this out Excell

=NORMSINV() Thresholds.xls

Page 10: Thresholds and ordinal data

Why do we rescale the data this way?

ConvenienceVariance always 1 Mean is always 0We can interpret the area under a

curve between two z-values as a probability or percentage

Page 11: Thresholds and ordinal data

Why do we rescale the data this way?

You could use other distributions but you would have to specify the fit function

Page 12: Thresholds and ordinal data

Threshold.R

Page 13: Thresholds and ordinal data

Threshold.R

Page 14: Thresholds and ordinal data

Threshold = .075 – Huh what?

Page 15: Thresholds and ordinal data

What about more than 2 categories?

Very similar We create a matrix containing the 1st

threshold and the displacements between subsequent matrices

We then add the 1st threshold and the displacement to obtain the subsequent thresholds

-3 31.20-1

2.2

Page 16: Thresholds and ordinal data

Mx Threshold Specification: 3+ Cat.

-3 31.20-1

2.2

1st threshold

Twin 1 Twin 2

increment

Page 17: Thresholds and ordinal data

Mx Threshold Model: Thresholds L*T /

1st threshold

Twin 1 Twin 2

increment

Mx Threshold Specification: 3+ Cat.

-3 31.20-1

2.2

Page 18: Thresholds and ordinal data

Mx Threshold Model: Thresholds L*T /

1st threshold

Twin 1 Twin 2

increment

2nd threshold

Mx Threshold Specification: 3+ Cat.

-3 31.20-1

2.2

Page 19: Thresholds and ordinal data

OpenMx code

-3 31.20-1

2.2

twin2zscore1- twin2zscore2 twin1zscore1- twin1zscore2

twin2zscore1 twin1zscore1T

The bounds stop the thresholds going ‘backwards’

ie they preserve the ordering of the data

Page 20: Thresholds and ordinal data

Lower bounds -3 -3 .0001 .0001

Page 21: Thresholds and ordinal data

OpenMx code

Page 22: Thresholds and ordinal data

OpenMx code

2nd threshold

Page 23: Thresholds and ordinal data

OpenMx code

2nd threshold

Page 24: Thresholds and ordinal data

Thresholds.xls

Try changing the values