# Bring polars into scope and import our dataset
import polars as pl
import altair as alt

df_cola = pl.read_csv("../datasets/cost-of-living-2018.csv")

df_cola

shape: (540, 7)

City	Cost of Living Index	Rent Index	Cost of Living Plus Rent Index	Groceries Index	Restaurant Price Index	Local Purchasing Power Index
str	f64	f64	f64	f64	f64	f64
"Hamilton, Bermuda"	145.43	110.87	128.76	143.47	158.75	112.26
"Zurich, Switzerland"	141.25	66.14	105.03	149.86	135.76	142.7
"Geneva, Switzerland"	134.83	71.7	104.38	138.98	129.74	130.96
"Basel, Switzerland"	130.68	49.68	91.61	127.54	127.22	139.01
"Bern, Switzerland"	128.03	43.57	87.3	132.7	119.48	112.71
…	…	…	…	…	…	…
"Kochi, India"	24.65	6.31	15.8	26.93	13.94	77.7
"Coimbatore, India"	24.61	5.35	15.32	25.23	15.21	53.23
"Alexandria, Egypt"	23.78	4.34	14.4	23.19	17.66	23.75
"Navi Mumbai, India"	23.44	6.25	15.15	24.02	14.14	111.99
"Thiruvananthapuram, India"	20.86	5.1	13.26	21.98	12.06	66.25

# Split out Country from City feature
df_cola = df_cola.with_columns(
    # The last element is the country
    pl.col("City").str.split(", ").list.last().alias("Country")
)


df_cola

shape: (540, 8)

City	Cost of Living Index	Rent Index	Cost of Living Plus Rent Index	Groceries Index	Restaurant Price Index	Local Purchasing Power Index	Country
str	f64	f64	f64	f64	f64	f64	str
"Hamilton, Bermuda"	145.43	110.87	128.76	143.47	158.75	112.26	"Bermuda"
"Zurich, Switzerland"	141.25	66.14	105.03	149.86	135.76	142.7	"Switzerland"
"Geneva, Switzerland"	134.83	71.7	104.38	138.98	129.74	130.96	"Switzerland"
"Basel, Switzerland"	130.68	49.68	91.61	127.54	127.22	139.01	"Switzerland"
"Bern, Switzerland"	128.03	43.57	87.3	132.7	119.48	112.71	"Switzerland"
…	…	…	…	…	…	…	…
"Kochi, India"	24.65	6.31	15.8	26.93	13.94	77.7	"India"
"Coimbatore, India"	24.61	5.35	15.32	25.23	15.21	53.23	"India"
"Alexandria, Egypt"	23.78	4.34	14.4	23.19	17.66	23.75	"Egypt"
"Navi Mumbai, India"	23.44	6.25	15.15	24.02	14.14	111.99	"India"
"Thiruvananthapuram, India"	20.86	5.1	13.26	21.98	12.06	66.25	"India"

# Calculate average cost of living by country and show the top 10
top_10_countries = (
    df_cola.group_by("Country")
    .agg(pl.mean("Cost of Living Index").alias("Avg Cost of Living"))
    .sort("Avg Cost of Living", descending=True)
    .head(10)
)

top_10_countries

shape: (10, 2)

Country	Avg Cost of Living
str	f64
"Bermuda"	145.43
"Switzerland"	130.088333
"Iceland"	123.78
"Norway"	115.5925
"Bahamas"	99.73
"Luxembourg"	95.37
"Denmark"	93.563333
"Singapore"	91.4
"Japan"	90.88
"South Korea"	87.56

# Barplot for top 10 cost of living
chart = (
    top_10_countries.plot.bar(
        x="Avg Cost of Living",
        y=alt.Y("Country", sort="-x")
    )
    .properties(
        title="Top 10 Most Expensive Countries"
    )
    .configure_axisY(
        title=None
    )
)

chart

# Calculate average purchasing power by country and show the top 10
top_10_power_countries = (
    df_cola.group_by("Country")
    .agg(pl.mean("Local Purchasing Power Index").alias("Avg Purchasing Power"))
    .sort("Avg Purchasing Power", descending=True)
    .head(10)
)

top_10_power_countries

shape: (10, 2)

Country	Avg Purchasing Power
str	f64
"Switzerland"	128.865
"Germany"	128.189444
"Luxembourg"	127.42
"Saudi Arabia"	127.0125
"United Arab Emirates"	126.09
"United States"	124.266591
"Australia"	123.193636
"Qatar"	120.86
"Finland"	116.763333
"Sweden"	116.754

# Barplot for top 10 purchasing power
chart = (
    top_10_power_countries.plot.bar(
        x="Avg Purchasing Power",
        y=alt.Y("Country", sort="-x")
    )
    .properties(
        title="Top 10 Countries by Purchasing Power"
    )
    .configure_axisY(
        title=None
    )
)

chart

# Scatter plot of Groceries vs. Restaurant Prices
chart = (
    df_cola.plot.point(
        x="Groceries Index",
        y="Restaurant Price Index",
        color=alt.Color("Country", legend=None),
        tooltip=["Groceries Index", "Restaurant Price Index", "City"]
    )
    .properties(
        width=600,
        height=400,
        title="Groceries Index vs. Restaurant Price Index"
    )
)

chart

# Scatter plot of Rent vs. Cost of Living
chart = (
    df_cola.plot.point(
        x="Rent Index",
        y="Cost of Living Index",
        color=alt.Color("Country", legend=None),
        tooltip=["Rent Index", "Cost of Living Index", "City"]
        )
        .properties(width=500, title="Rent vs. Cost of Living")
        .configure_scale(zero=False)
)

chart

# Line plot showing cost of living plus rent index across city
df_sorted_cities = df_cola.sort("Cost of Living Plus Rent Index")

chart = (
    df_sorted_cities.plot.line(
        x=alt.X("City", sort=None),
        y="Cost of Living Plus Rent Index",
        tooltip=["City", "Cost of Living Plus Rent Index"]
    )
    .properties(
        width=1200,
        height=500,
        title="Cost of Living Plus Rent Index Across All Cities (Sorted)"
    )
    .configure_axisX(
        title=None,
        labels=False,
        ticks=False
    )
)

chart