In this world of selfies I make data selfies. I documented my daily routine and weekly goals for 38 weeks and how I have some data to make conclusions on.
According to my own observations of myself the earlier I wake up and start working, the more weekly goals I achieve. Starting the work at 10 a.m. yields better results than at 9 a.m., but there were only 3 weeks when I woke up so late on average and even those were affected by a single day when I slept until noon after some kind of all-night activity.
When I wake up early my body and mind are ready to work at once – I’m definitely not an owl.
What: Weekly average time of starting to work, and the proportion of weekly goals achieved. Time is binned into intervals by 1 hour, and the proportion is averaged. When: 38 weeks during 2019 and 2020 Source: self-observation
My real goal here was to draw really “creative” chart even if it is hard to read.
They’re doing well on the internet. China’s Tencent and Alibaba are among the top 10 for 4 years already.
What: Non-USA companies that are in the top 10 public traded companies by market capitalization every year. Saudi Aramco is not included, because only a small fraction of its shares are traded publicly. When: 1997-2020Q2 Where: I hope it’s the whole world except the USA. Source: https://en.wikipedia.org/wiki/List_of_public_corporations_by_market_capitalization
What: Top 10 public traded companies by market capitalization every year. Saudi Aramco is not included, because only a small fraction of its shares are traded publicly. When: 1997-2020Q2 Where: I hope it’s the whole world. Source: https://en.wikipedia.org/wiki/List_of_public_corporations_by_market_capitalization
Every drop in GDP is followed by a decline in real estate prices or at least a significant slow down in growth with the most obvious case at the beginning of ’80s.
What: Moving average of quarterly USA GDP growth (GDP in billions of chained 2012 dollars, seasonally adjusted) and moving average of quarterly median sales price of houses sold in the United States. When: 1964Q1 – 2020Q1 Where: USA only. Source: FRED
Įsivaizduok – nori nuvažiuoti iš Vilniaus į Kauną kelioms valandoms ir tada grįžti. Sužiūrėti traukinių tvarkaraščius tokiai kelionei nėra sudėtinga, bet visgi reikia daryti dvi paieškas ir nepasiklysti tarp išvykimo-atvykimo laikų.
Argi nebūtų patogu, jei kelionės į abi puses būtų išdėliotos vienoje schemoje?
Įdomus faktas – 6:36 iš Kauno išvykstantis traukinys pakeliui sutinka 4 priešinga kryptimi važiuojančius traukinius.
Often when the market turns red there is at least a dip in GDP growth. However, there are cases when this does not happen.
What: Moving average of quarterly S&P500 index growth and moving average of quarterly USA GDP growth (GDP in billions of chained 2012 dollars, seasonally adjusted) When: 1948Q1 – 2020Q1 Where: USA only, because S&P is populated by mostly American companies. Source: Yahoo Finance for S&P500 and FRED for GDP.
I saw that Equatorial Guinea, the African country that increased its GDP per capita the most during 1980-2018 had done this due to newly found oil. I saw that Gabon, the African country that decreased its GDP per capita the most during the same period, had done this due to diminishing its oil reserves. So I thought, the same applies to the most of rapidly growing (or contracting) African economies, but that is not true (except for Libya).
The sources of growth might not be very sustainable just like oil, but finding them requires a deeper analysis of every separate country.
What: GDP per capita divided into oil rents, rents from other natural resources and GDP from other sources. Grey bars indicate full GDP value with the unknown division. When: From 1980 till 2018. Not all countries had a full range of data. No country had its GDP divided for 2018. Where: Countries of the African continent with the biggest GDP per capita growth during 1980-2018 (top 8) and biggest fall (bottom 4, except South Sudan, which had very short data range) Source: WB
Many people still have doubts about whether the lower emissions of driving an electric car outweighs the additional impact on the environment caused by manufacturing the batteries.
Two studies (the latter is more trustworthy than the former) show that electric cars do help to reduce the CO2 emission. But how much – it depends. If the batteries are produced in a country where the industry is polluting more (China), reductions will be lower. If the car is driven in a country where electricity is produced by burning things (like Estonia or Poland) the reductions will be lower or even negative. A special case is Japan, where cars are so fuel-efficient, that electric cars even raise emissions.
So, YES, electric cars are more often better than not.
What: Top = Lifetime emissions of cars in tonnes of CO2. Bottom = Lifetime emissions of cars in grams of CO2 per 1 kilometer. When: Top = Estimate for 2020, Bottom = Estimate for 2030 with “current technological trajectory” scenario. Where: Top = EU countries, Bottom = selected countries of the world. Source: Top = European Federation for Transport and Environment, Bottom = Knobloch, F., Hanssen, S., Lam, A. et al. Net emission reductions from electric cars and heat pumps in 59 world regions over time. Nat Sustain (2020). https://doi.org/10.1038/s41893-020-0488-7
I have seen lots of images where all R colours are arranged alphabetically or in some other strange order, including all grays and greys.
Here colours are grouped by name, sorted according to their hue and arranged by lightness inside a group, and saturation among groups. The table is not perfect since all the positions were calculated – my “colour recognition algorithm” was not perfect. (This is how “mediumspringgreen” finished among cyans.) But it is definitely better than most tables.
We spend more and more money on movies even in times of television and the internet. So, not a big surprise, that more and more movies are being made.
Median rating of all movies in the IMDB database is declining. Is it a sign of movies getting worse? No, its a sign of more movies being made, when this art is accessible to not only exceptional talents but mediocre talents as well. Anyway, who watches 2000 movies a year? Let’s become picky.
Let’s pick the most popular 200 movies each year and calculate their median. Now there is a stable trend – median popular movie usually has a rating of just around 6.4. (Movie popularity is measured in the number of ratings on IMDB). That’s good.
Now let’s take 200 top-rated movies each year with at least 25 000 ratings. A sharp decline becomes visible! And before the 80s not many such movies existed at all. So, if you like only top-rated popular movies, you might experience deterioration in their quality. Sequels of Transformers are getting worse and worse.
However there are great lesser-known movies, especially those made in non-English speaking countries, so they get less attention and therefore fewer ratings. I’d subjectively say that a good lesser-known movie has at least 4000 ratings. If we pick 200 best such movies, we get a different view – often a median movie will have a rating higher than 7, and no permanent downward trend is present. Great!
What: Movie earnings in USD billion adjusted for inflation. The number of movies – with more than 100, 4000 and 25 000 ratings. IMDB movie ratings indicate the median rating calculated for every year, and a trend line fitted using a generalized additive model. All movies – those which have at least 100 ratings Most popular movies – 200 most rated but having at least 100 ratings (because there were times when less than 200 movies were made). Top-rated popular movies – 200 top-rated movies from those having at least 25 000 ratings. Top-rated movies for cinephiles – 200 top-rated movies from those having at least 4000 ratings. If less than 40 movies left remaining after filtering – no rating calculated for that year. When: 1968 – 2019 Where: Movie earnings represent earnings around the globe. Source: IMDB for ratings and number of movies the-numbers.com for movie earnings. WB and IMF for inflation used to adjust earnings.
Based on how many average baskets of goods one can buy for an average salary while living alone and without children – yes, absolutely. There were wiggles after the crisis, but the overall trend is upwards – people can buy more and more stuff because salaries are increasing faster than prices. The only sad exception in the period analyzed is Slovakia.
What: Annual net earnings divided by price level index which result in the number of standard baskets of goods one can buy. The average wage is “annual net earnings of a full-time single worker without children earning an average wage” measured in Purchase Power Standards. Price level index is made to be 100 in 2020 for the EU as a whole for a standardized basket of goods. It is tailored for cross-country comparison. Data for OECD countries was estimated using “Price level ratio of PPP conversion factor (GDP) to market exchange rate” (provided by WB, very accurate estimates), and “Average annual wage in 2018 constant prices at 2018 USD PPPs” (provided by OECD, not very accurate estimates) When: 2000 – 2018 Where: 39 countries from the EU and OECD. Source: EUROSTAT, OECD, WB
Based on how many average baskets of goods one can buy for an average salary while living alone and without children – Poland takes the lead followed by Germany. For some time the United States seemed the best country for that. However in a few decades, things can change drastically – the Slovak Republic once at the top, now is somewhere at the bottom, and Poland itself climbed a long way up – the time between 2009 and 2010 was a mess.
It is very difficult to find time-series data about the average salary, so only the EU and OECD countries are included in this chart. It would be interesting to see how cheap is Thailand for Thai people!
What: Average wage divided by price level index which results in the number of standard baskets of goods one can buy. The average wage is “annual net earnings of a full-time single worker without children earning an average wage” measured in Purchase Power Standards. The price level index is made to be 100 in 2020 for the EU as a whole for a standardized basket of goods. It is tailored for cross-country comparison. Data for OECD countries was estimated using “Price level ratio of PPP conversion factor (GDP) to market exchange rate” (provided by WB, very accurate estimates), and “Average annual wage in 2018 constant prices at 2018 USD PPPs” (provided by OECD, not very accurate estimates) When: 2000 – 2018 Where: 39 countries of the world Source: EUROSTAT, OECD, WB
The good news is that we are really reducing poverty. Fewer and fewer people live for 1,90 dollars a day, fewer and fewer experience hunger, fewer and fewer experience struggle to get water.
The sad news is the realization that behind nice world trends there are still countries where one-third of the total population are poor, experience hunger and struggle to get water. And even worse – while over 40% of people in low-income countries don’t have access to BASIC drinking water service, 73% don’t have access to SAFE drinking water service.
But we’re reducing. There is still much to achieve and let’s hope for the best.
What: Very poor people = Poverty headcount ratio at $1.90 a day (2011 PPP) (% of population), Hungry people = Prevalence of undernourishment (% of population), People not having water = People not using at least basic drinking water services (% of population) When: Improvements by income class are shown between 2000 and 2017 with possible deviations a year or two for different category combinations. Where: Total world + the world divided into 4 income groups by the World Bank. Source: WB
Livestock. All this data is very unreliable, especially for animals living in the wild, but some rough estimates were made, and we know that there are probably 30 times more mammals kept as livestock than wild land mammals, and 2 times more birds kept as livestock than wild birds. Also, there are as many cats and dogs as animals in the wild.
These “quantities” are actually a total mass of carbon, which according to the Source make roughly 15% of total body mass. So, there might be lots of chicken, but they don’t weight as much as cows. The same with little wild animals – there might be more rats than cows.
Also, some types of animals were not taken into account, for example, reptiles, arthropods, sea mammals. I decided to concentrate on mammals and birds living above the ground. I’d include reptiles, but there are no reliable estimations about them.
What: Mass of carbon inside mammals and birds. When: Data were taken from a study published in 2018, which used data from FAO. So, I guess it’s not very old. Where: Land and above. Source: Bar-On, Yinon M et al. “The biomass distribution on Earth.” Proceedings of the National Academy of Sciences of the United States of America vol. 115,25 (2018). Also, https://www.worldatlas.com used to estimate the mass of cats and dogs which were not included in the paper above.