greeners 1.5.3

High-performance econometrics with R/Python formulas. Two-Way Clustering, Marginal Effects (AME/MEM), HC1-4, IV Predictions, Categorical C(var), Polynomial I(x^2), Interactions, Diagnostics. OLS, IV/2SLS, DiD, Logit/Probit, Panel (FE/RE), Time Series (VAR/VECM), Quantile!
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146

// Example demonstrating various ways to load data into a DataFrame
//
// This example shows:
// 1. CSV from local file
// 2. CSV from URL
// 3. JSON from local file
// 4. JSON from URL
// 5. Builder pattern
// 6. Direct construction from HashMap

use greeners::{DataFrame, Formula};
use indexmap::IndexMap;
use ndarray::Array1;

fn main() {
    println!("=== DataFrame Loading Examples ===\n");

    // Method 1: Direct construction from HashMap
    println!("1. Direct construction from HashMap:");
    let mut data = IndexMap::new();
    data.insert("x".to_string(), Array1::from(vec![1.0, 2.0, 3.0, 4.0]));
    data.insert("y".to_string(), Array1::from(vec![2.0, 4.0, 6.0, 8.0]));

    let df1 = DataFrame::new(data).expect("Failed to create DataFrame");
    println!("   Loaded {} rows x {} columns", df1.n_rows(), df1.n_cols());
    println!("   Columns: {:?}\n", df1.column_names());

    // Method 2: Builder pattern (most convenient for simple cases)
    println!("2. Builder pattern:");
    let df2 = DataFrame::builder()
        .add_column("price", vec![100.0, 150.0, 200.0])
        .add_column("quantity", vec![10.0, 20.0, 15.0])
        .add_column("profit", vec![50.0, 75.0, 100.0])
        .build()
        .expect("Failed to build DataFrame");

    println!("   Loaded {} rows x {} columns", df2.n_rows(), df2.n_cols());
    println!("   Columns: {:?}\n", df2.column_names());

    // Method 3: CSV from local file
    println!("3. CSV from local file:");
    println!("   Creating a test CSV file...");

    // Create a test CSV file
    use std::fs::File;
    use std::io::Write;
    let mut file = File::create("test_data.csv").expect("Failed to create file");
    writeln!(file, "income,education,experience").unwrap();
    writeln!(file, "50000,16,5").unwrap();
    writeln!(file, "60000,18,7").unwrap();
    writeln!(file, "55000,16,6").unwrap();
    writeln!(file, "70000,20,10").unwrap();

    let df3 = DataFrame::from_csv("test_data.csv").expect("Failed to read CSV");
    println!("   Loaded {} rows x {} columns", df3.n_rows(), df3.n_cols());
    println!("   Columns: {:?}\n", df3.column_names());

    // Method 4: JSON from local file (column-oriented)
    println!("4. JSON from local file (column-oriented):");
    println!("   Creating a test JSON file...");

    let mut json_file = File::create("test_data_columns.json").expect("Failed to create file");
    writeln!(json_file, "{{").unwrap();
    writeln!(json_file, r#"  "age": [25, 30, 35, 40],"#).unwrap();
    writeln!(json_file, r#"  "salary": [40000, 50000, 60000, 70000],"#).unwrap();
    writeln!(json_file, r#"  "years_exp": [2, 5, 8, 12]"#).unwrap();
    writeln!(json_file, "}}").unwrap();

    let df4 = DataFrame::from_json("test_data_columns.json").expect("Failed to read JSON");
    println!("   Loaded {} rows x {} columns", df4.n_rows(), df4.n_cols());
    println!("   Columns: {:?}\n", df4.column_names());

    // Method 5: JSON from local file (record-oriented)
    println!("5. JSON from local file (record-oriented):");
    println!("   Creating a test JSON file...");

    let mut json_file2 = File::create("test_data_records.json").expect("Failed to create file");
    writeln!(json_file2, "[").unwrap();
    writeln!(json_file2, r#"  {{"height": 170.0, "weight": 70.0}},"#).unwrap();
    writeln!(json_file2, r#"  {{"height": 175.0, "weight": 75.0}},"#).unwrap();
    writeln!(json_file2, r#"  {{"height": 180.0, "weight": 80.0}}"#).unwrap();
    writeln!(json_file2, "]").unwrap();

    let df5 = DataFrame::from_json("test_data_records.json").expect("Failed to read JSON");
    println!("   Loaded {} rows x {} columns", df5.n_rows(), df5.n_cols());
    println!("   Columns: {:?}\n", df5.column_names());

    // Method 6: CSV from URL (example - requires internet connection)
    println!("6. CSV from URL:");
    println!("   Note: This requires internet connection and a valid CSV URL");
    println!("   Example usage:");
    println!(
        r#"   let df = DataFrame::from_csv_url("https://github.com/sheep-farm/Greeners/blob/main/data.csv")?;"#
    );
    println!();

    // Uncomment to test with a real URL:
    let df6 = DataFrame::from_csv_url(
        "https://raw.githubusercontent.com/sheep-farm/Greeners/refs/heads/main/data.csv",
    )
    .unwrap();
    println!("   Loaded {} rows x {} columns", df6.n_rows(), df6.n_cols());

    // Method 7: JSON from URL (example - requires internet connection)
    println!("7. JSON from URL:");
    println!("   Note: This requires internet connection and a valid JSON URL");
    println!("   Example usage:");
    println!(
        r#"   let df = DataFrame::from_json_url("https://raw.githubusercontent.com/sheep-farm/Greeners/refs/heads/main/data.json")?;"#
    );
    println!();

    // Bonus: Using the DataFrame with formulas
    println!("\n=== Bonus: Using DataFrame with OLS regression ===\n");

    let df_regression = DataFrame::builder()
        .add_column("wage", vec![30000.0, 40000.0, 50000.0, 60000.0, 70000.0])
        .add_column("education", vec![12.0, 14.0, 16.0, 18.0, 20.0])
        .add_column("experience", vec![2.0, 4.0, 6.0, 8.0, 10.0])
        .build()
        .expect("Failed to build DataFrame");

    let formula = Formula::parse("wage ~ education + experience").expect("Failed to parse formula");
    let (y, x) = df_regression
        .to_design_matrix(&formula)
        .expect("Failed to create design matrix");

    println!(
        "Design matrix shape: {} rows x {} columns",
        x.nrows(),
        x.ncols()
    );
    println!("Response vector length: {}", y.len());

    // You can now use this with OLS or any other model:
    // use greeners::OLS;
    // let model = OLS::fit(y, x)?;
    // model.summary();

    // Clean up test files
    println!("\n=== Cleaning up test files ===");
    std::fs::remove_file("test_data.csv").ok();
    std::fs::remove_file("test_data_columns.json").ok();
    std::fs::remove_file("test_data_records.json").ok();
    println!("Test files removed.");
}