Audio Transcript Auto-generated
- 00:01 - 00:01
mhm.
- 00:02 - 00:05
Hi everyone, this is Christina Kris Janowski.
- 00:05 - 00:05
Today,
- 00:05 - 00:07
I'm going to talk to you a little bit about
- 00:07 - 00:14
average global carbon dioxide emissions from 1984 until 2018.
- 00:14 - 00:16
Um
- 00:17 - 00:20
Right here, you can see that my data source,
- 00:20 - 00:22
I obtained all my raw data from noAA dot gov.
- 00:22 - 00:29
Um The source obtained um was obtained from a population.
- 00:29 - 00:33
We created a random sample by using google sheets that ran
- 00:33 - 00:37
between function and randomly selected a sample size of 35.
- 00:37 - 00:41
This actually minimized the possibility for bias.
- 00:41 - 00:47
Um The population is the average global carbon dioxide emissions.
- 00:48 - 00:51
Um so we chose this topic because we believe that the
- 00:51 - 00:56
carbon dioxide emissions are higher than they were in 1980.
- 00:56 - 00:58
Uh if you look at the spreadsheets,
- 00:58 - 01:01
you can see that the first two columns are raw data.
- 01:01 - 01:06
The third column um is the random sample of
- 01:06 - 01:09
road numbers in the last two columns were also generated
- 01:10 - 01:12
based on the raw data road numbers.
- 01:12 - 01:17
Um It's important for me to note um as stated by Noah,
- 01:17 - 01:22
the uncertainty in the global annual mean is estimated using a
- 01:22 - 01:28
monte monte Carlo technique that computes 100 global annual averages.
- 01:28 - 01:33
So CO two is expressed as a mole fraction in dry air
- 01:33 - 01:36
um abbreviated as parts per million.
- 01:37 - 01:40
Uh My summary here,
- 01:40 - 01:44
um it's really hard to read. Let me see if I can blow that up. So, the hissed a gram.
- 01:44 - 01:46
This system graham here shows
- 01:47 - 01:52
um that the distribution is um not normal.
- 01:52 - 01:57
Um you can see that there's no bell curve and there's just a slight um
- 01:57 - 02:01
right skew half of the data falls within
- 02:01 - 02:05
approximately one standard deviation below the mean,
- 02:05 - 02:08
and then you can kind of see as it goes to the right.
- 02:09 - 02:11
Um The rest of the data is just
- 02:11 - 02:15
kind of spread out approximately two standard deviations um
- 02:15 - 02:17
above me.
- 02:18 - 02:22
Um You can see here, I decided to do a scatter plot.
- 02:22 - 02:24
So the scatter plot,
- 02:24 - 02:27
um you can see that it actually shows an
- 02:27 - 02:32
upward trend in carbon dioxide emissions over time.
- 02:33 - 02:33
Yeah.
- 02:34 - 02:41
Um so some of the other the data that we got on that summary sheet that
- 02:41 - 02:47
second um tab I kind of figured this was some of them more the most important
- 02:47 - 02:52
um of that data, you can see um Well, it doesn't show it here,
- 02:52 - 02:57
but the average deviation um for this data set is 22.
- 02:57 - 03:03
Um That's the deviation. The standard deviation. Um the five number summary.
- 03:03 - 03:05
So that's going to be a minimum cute one,
- 03:06 - 03:09
The median or Q two, Q 3, and then the max
- 03:09 - 03:13
um you're going to see that it indicates no outlier.
- 03:13 - 03:17
So, to clarify if the data falls below the lower fence or above the upper fence,
- 03:17 - 03:20
that data would be considered an outlier.
- 03:20 - 03:23
So the lower fence is much lower than minimum value,
- 03:23 - 03:29
and then the upper friends is much higher than the maximum number, so hence no
- 03:30 - 03:31
no outliers.
- 03:35 - 03:39
Um I decided to do the confidence interval
- 03:39 - 03:42
um firming and the hypothesis tests for means.
- 03:42 - 03:47
So we were able to conclude that 95 were 95% confident that the
- 03:47 - 03:50
actual population mean for carbon dioxide emissions
- 03:50 - 03:55
between 1984 and 2018 is between 364
- 03:55 - 03:59
Um .26 and 3 79.2.
- 03:59 - 04:04
So I remember that the point estimate is the midpoint of the confidence interval.
- 04:05 - 04:09
Um The hypothesis tests from the main previous data indicated that
- 04:09 - 04:15
the carbon dioxide emissions averaged 300 fold 350 moles in 1980.
- 04:15 - 04:16
So given the climate crisis,
- 04:16 - 04:20
it seems reasonable that the value may no longer be value
- 04:20 - 04:26
be valid. Um We assumed in the null hypothesis that the status quo
- 04:27 - 04:31
um of 350 moles. The alternate hypothesis is
- 04:32 - 04:35
basically that the average is now greater than 3 50.
- 04:36 - 04:39
You can see on the spreadsheet,
- 04:39 - 04:44
the one tailed test indicates a critical value of 1.69 with the right tail.
- 04:44 - 04:50
Um and you can kind of see that again on the on that hissed a gram or test statistics.
- 04:50 - 04:55
Uh the statistic falls far to the right and our p value is way less than the alpha.
- 04:56 - 04:59
So therefore we can we reject the
- 04:59 - 05:02
null hypothesis and concluded the actual population,
- 05:02 - 05:04
I mean is greater than 3 50.
- 05:04 - 05:06
The confidence interval,
- 05:06 - 05:13
you can also see the that confidence interval of 3 64 and 3 79 does not include 3 50.
- 05:13 - 05:17
So, again, this confirms the rejection of the null hypothesis.
- 05:18 - 05:22
So, um just to kind of wrap up the presentation, you can kind of
- 05:22 - 05:26
So in my conclusion here you can see that our data analysis on the population mean of
- 05:26 - 05:31
global global carbon dioxide emissions from 1984 to 2008
- 05:31 - 05:34
indicated a slight right scoob and zero liars.
- 05:35 - 05:41
Um So a couple of the questions um or what are the weakness?
- 05:41 - 05:43
Uh the weaknesses in your experiment.
- 05:43 - 05:48
So I would say the weakness again is just uh you know the
- 05:48 - 05:52
ran between function um that we have some of that repeat data.
- 05:52 - 05:56
How can you correct these? I think the
- 05:56 - 05:59
the best way to do that would just we would um
- 06:00 - 06:03
We would find an alternative way to create a random sample.
- 06:03 - 06:09
We could also add years um prior to 1984 and then just some more current data.
- 06:09 - 06:14
How would you further your analysis? What I would say would just be
- 06:15 - 06:15
uh
- 06:16 - 06:21
to do a regression or a time series analysis with either regression or time series.
- 06:21 - 06:27
We would be able to predict co two emissions um in the coming years.
- 06:28 - 06:32
Um So that's my project.
- 06:32 - 06:37
Uh If you have any questions feel free to ask in the comment section,
- 06:38 - 06:40
um Thank you for listening and happy holidays