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
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227

use serde_json;
use serde_json::Value;

pub struct RegressionDataSetTempEntry {
    price: f64,
    price_small: f64,
    begin_timestamp: f64,
    timestamp: f64,
    timestamp_small: f64,
    time_value: f64,
    time_square: f64,
}

pub struct RegressionDatasetFullEntry {
    price: f64,
    price_small: f64,
    begin_timestamp: f64,
    timestamp: f64,
    timestamp_small: f64,
    time_value: f64,
    time_square: f64,
    sum_price_small: f64,
    sum_time: f64,
    sum_time_value: f64,
    sum_time_square: f64,
    regression_a: f64,
    regression_b: f64,
    regression_value: f64,
    regression_a_abs: f64,
    regression_b_half: f64,
}

pub fn calculate_initial_regression(price_dataset: &Value) -> Vec<RegressionDatasetFullEntry> {
    let temp_dataset: Vec<RegressionDataSetTempEntry> = make_basic_regression_dataset(price_dataset, None);
    let mut dataset: Vec<RegressionDatasetFullEntry> = Vec::new();

    let mut sum_price_small: f64 = 0.0;
    let mut sum_time: f64 = 0.0;
    let mut sum_time_value: f64 = 0.0;
    let mut sum_time_square: f64 = 0.0;

    for entry in temp_dataset.iter() {
        sum_price_small += entry.price_small;
        sum_time += entry.timestamp_small;
        sum_time_value += entry.time_value;
        sum_time_square += entry.time_square;
    }

    let regression_a: f64 = calculate_regression_a(sum_price_small, sum_time_square, sum_time, sum_time_value, temp_dataset.len() as f64);
    let regression_b: f64 = calculate_regression_b(sum_price_small, sum_time_square, sum_time, sum_time_value, temp_dataset.len() as f64);
    let regression_value: f64 = calculate_regression_value(regression_a, regression_b, temp_dataset[temp_dataset.len() - 1].timestamp_small);
    let regression_a_abs: f64 = calculate_regression_a_abs(regression_a, regression_b, temp_dataset[0].begin_timestamp);
    let regression_b_half: f64 = calculate_initial_regression_b_half(regression_b);

    for entry in temp_dataset.iter() {
        dataset.push(RegressionDatasetFullEntry {
            price: entry.price,
            price_small: entry.price_small,
            begin_timestamp: entry.begin_timestamp,
            timestamp: entry.timestamp,
            timestamp_small: entry.timestamp_small,
            time_value: entry.time_value,
            time_square: entry.time_square,
            sum_price_small,
            sum_time,
            sum_time_value,
            sum_time_square,
            regression_a,
            regression_b,
            regression_value,
            regression_a_abs,
            regression_b_half,
        })
    }

    dataset
}

pub fn update_regression_dataset(mut regression_dataset: Vec<RegressionDatasetFullEntry>, price_dataset: &Value, regression_length_in_seconds: f64) -> Vec<RegressionDatasetFullEntry>{
    let temp_dataset: Vec<RegressionDataSetTempEntry> = make_basic_regression_dataset(price_dataset, Some(regression_dataset[regression_dataset.len() - 1].begin_timestamp));

    for i in 0..temp_dataset.len() {
        let mut sum_time: f64 = regression_dataset[regression_dataset.len() - 1].sum_time + temp_dataset[i].timestamp_small;
        let mut sum_price_small: f64 = regression_dataset[regression_dataset.len() - 1].sum_price_small + temp_dataset[i].price_small;
        let mut sum_time_value: f64 = regression_dataset[regression_dataset.len() - 1].sum_time_value + temp_dataset[i].time_value;
        let mut sum_time_square: f64 = regression_dataset[regression_dataset.len() - 1].sum_time_square + temp_dataset[i].time_square;

        let mut current_begin_timestamp: f64 = temp_dataset[0].begin_timestamp;
        let mut entry_to_be_deleted: Vec<usize> = Vec::new();
        for (usize, entry) in regression_dataset.iter().enumerate() {
            if temp_dataset[i].timestamp - entry.timestamp > regression_length_in_seconds {
                sum_time -= entry.timestamp_small;
                sum_price_small -= entry.price_small;
                sum_time_value -= entry.time_value;
                sum_time_square -= entry.time_square;
                entry_to_be_deleted.push(usize);
            } else {
                break;
            }
        }

        // Sort in descending order to avoid issues with shifting elements and remove the elements from the Vec
        entry_to_be_deleted.sort_by(|a, b| b.cmp(a));
        for &index in &entry_to_be_deleted {
            if index < regression_dataset.len() {
                regression_dataset.remove(index);
            } else {
                println!("Index {} out of bounds", index);
            }
        }

        let regression_a: f64 = calculate_regression_a(sum_price_small, sum_time_square, sum_time, sum_time_value, regression_dataset.len() as f64 + 1.0);
        let regression_b: f64 = calculate_regression_b(sum_price_small, sum_time_square, sum_time, sum_time_value, regression_dataset.len() as f64 + 1.0);
        let regression_value: f64 = calculate_regression_value(regression_a, regression_b, temp_dataset[i].timestamp_small);
        let regression_a_abs: f64 = calculate_regression_a_abs(regression_a, regression_b, temp_dataset[i].begin_timestamp);


        let check_timestamp: f64 = temp_dataset[i].timestamp - (0.5 * regression_length_in_seconds);
        let mut regression_b_half: f64 = 123456789.12;
        for entry in regression_dataset.iter() {
            regression_b_half = entry.regression_b;
            if entry.timestamp > check_timestamp {
                break;
            }
        }

        if regression_b_half == 123456789.12 {
            panic!("ILLEGAL");
        }

        regression_dataset.push(
            RegressionDatasetFullEntry {
                price: temp_dataset[i].price,
                price_small: temp_dataset[i].price_small,
                begin_timestamp: current_begin_timestamp,
                timestamp: temp_dataset[i].timestamp,
                timestamp_small: temp_dataset[i].timestamp_small,
                time_value: temp_dataset[i].time_value,
                time_square: temp_dataset[i].time_square,
                sum_price_small,
                sum_time,
                sum_time_value,
                sum_time_square,
                regression_a,
                regression_b,
                regression_value,
                regression_a_abs,
                regression_b_half,
            }
        )
    }

    regression_dataset
}


fn make_basic_regression_dataset(data_set_raw: &Value, begin_timestamp: Option<f64>) -> Vec<RegressionDataSetTempEntry> {
    let mut result: Vec<RegressionDataSetTempEntry> = Vec::new();
    let mut begin_timestamp: f64 = begin_timestamp.unwrap_or(-0.1);
    if let Value::Array(data_array) = data_set_raw {
        for element in data_array {
            let price: f64 = get_price_as_f64(&element);
            let timestamp: f64 = get_timestamp_as_f64(&element);
            if begin_timestamp == -0.1 {
                begin_timestamp = timestamp;
            }
            let price_small: f64 = price / 1000.0;
            let timestamp_small: f64 = (timestamp - begin_timestamp) / 100000.0;
            let time_value: f64 = timestamp_small * price_small;
            let time_square: f64 = f64::powf(timestamp_small, 2.0);
            result.push(RegressionDataSetTempEntry {
                price,
                price_small,
                begin_timestamp,
                timestamp,
                timestamp_small,
                time_value,
                time_square,
            });
        }
    }
    result
}

fn get_timestamp_as_f64(element: &Value) -> f64 {
    let timestamp_str: &str = element["timestamp"].as_str().unwrap();
    match timestamp_str.parse::<f64>() {
        Ok(parsed_timestamp) => {
            parsed_timestamp
        }
        _ => { panic!("error parsing timestamp from element") }
    }
}

fn get_price_as_f64(element: &Value) -> f64 {
    let price_str: &str = element["price"].as_str().unwrap();
    match price_str.parse::<f64>() {
        Ok(parsed_price) => {
            parsed_price
        }
        _ => { panic!("error parsing price from element") }
    }
}

fn calculate_regression_a(
    sum_price_small: f64, sum_time_square: f64, sum_time: f64, sum_time_value: f64, regression_len: f64,
) -> f64 {
    (((sum_price_small * sum_time_square) - (sum_time * sum_time_value)) / (regression_len * sum_time_square - f64::powf(sum_time, 2.0))) * 1000.0
}

fn calculate_regression_b(
    sum_price_small: f64, sum_time_square: f64, sum_time: f64, sum_time_value: f64, regression_len: f64,
) -> f64 {
    (((regression_len * sum_time_value) - (sum_time * sum_price_small)) / (regression_len * sum_time_square - f64::powf(sum_time, 2.0)) ) / 100.0
}

fn calculate_regression_value(regression_a: f64, regression_b: f64, time_stamp: f64) -> f64 {
    regression_a + ((regression_b * time_stamp) * 100000.0)
}

fn calculate_regression_a_abs(regression_a: f64, regression_b: f64, begin_timestamp: f64) -> f64 {
    regression_a - ((begin_timestamp / 100000.0) * regression_b)
}

fn calculate_initial_regression_b_half(regression_b: f64) -> f64 {
    regression_b
}