WEBVTT

00:00:05.050 --> 00:00:07.760
Let us introduce
the error measures

00:00:07.760 --> 00:00:11.360
we used in building
the analytics models.

00:00:11.360 --> 00:00:16.430
We of course used R squared,
but we also used other measures.

00:00:19.550 --> 00:00:22.570
Next measure, the
so-called "penalty error,"

00:00:22.570 --> 00:00:26.420
is motivated by the fact
that if you classify

00:00:26.420 --> 00:00:29.600
a very high-risk patient
as a low-risk patient,

00:00:29.600 --> 00:00:32.299
this is more costly
than the reverse,

00:00:32.299 --> 00:00:34.820
namely classifying
a low-risk patient

00:00:34.820 --> 00:00:35.980
as a very high-risk patient.

00:00:39.410 --> 00:00:44.200
Motivated by this, we
developed a penalty error.

00:00:44.200 --> 00:00:48.050
And the idea is to use
asymmetric penalties.

00:00:48.050 --> 00:00:51.380
The graph here--
so it's a matrix--

00:00:51.380 --> 00:00:54.880
where this is the outcome
and this is the forecast.

00:00:54.880 --> 00:01:03.260
For example, whenever we
classify a low-risk patient

00:01:03.260 --> 00:01:06.820
as high-risk, we
pay a penalty of 2,

00:01:06.820 --> 00:01:12.720
which is a difference of 3 minus
1, the difference in the error.

00:01:12.720 --> 00:01:21.250
But inversely, when you
classify a bracket 3 patient

00:01:21.250 --> 00:01:24.340
as bracket 1 patient,
this is double.

00:01:24.340 --> 00:01:27.150
The cost-- the penalty--
is double the amount.

00:01:27.150 --> 00:01:34.340
So you observe that the off
diagonal penalties are double

00:01:34.340 --> 00:01:36.860
the corresponding penalties
in the lower diagonal.

00:01:39.789 --> 00:01:46.860
To judge the quality of the
analytics models we developed,

00:01:46.860 --> 00:01:48.440
we compare it with a baseline.

00:01:48.440 --> 00:01:50.100
And the baseline is
to simply predict

00:01:50.100 --> 00:01:52.150
that the cost in
the next "period"

00:01:52.150 --> 00:01:54.850
will be the cost in
the current period.

00:01:54.850 --> 00:01:59.170
We have observed that as far
as identification of brackets

00:01:59.170 --> 00:02:03.120
is concerned, the
accuracy was 75%.

00:02:03.120 --> 00:02:10.830
So namely, whenever we predict
that the risk is bracket 3--

00:02:10.830 --> 00:02:14.410
indeed it is bracket 3--
this happens 75% of the time,

00:02:14.410 --> 00:02:16.600
and the penalty error--
the average penalty

00:02:16.600 --> 00:02:20.190
error of the
baseline-- was 0.56.